JP2007524342A - B12-dependent dehydratase with improved reaction rate - Google Patents

Abstract

By providing an array of B _12-dependent dehydratase with improved reaction kinetics, the rate of suicide inactivation of enzymes in glycerol and 1,3-propanediol presence is reduced. The enzyme is created using error-prone PCR and oligonucleotide-induced mutagenesis targeted to the DhaB1 gene encoding the α-subunit of glycerol dehydratase. Variants with improved reaction rates are rapidly identified using the high-throughput assay also disclosed herein.

Description

The present invention relates to the use of molecular biology and B ₁₂ dependent dehydratase to produce 1,3-propanediol. More specifically, creating a B ₁₂ dependent dehydratase reaction rate was improved to decrease enzyme inactivation rate, describes a method of selecting.

1,3-propanediol has utility in some applications, including starting materials for making polyesters, polyethers, and polyurethanes. Methods for producing 1,3-propanediol include both traditional chemical and biological routes. A biological method for producing 1,3-propanediol has recently been described (Non-Patent Document 1). Biological production of 1,3-propanediol requires glycerol as a substrate for a two-step sequential reaction. First, a dehydratase (typically a coenzyme B _12- dependent dehydratase) converts glycerol to the intermediate 3-hydroxypropionaldehyde (3-HP). Next, 3-HP is reduced to 1,3-propanediol by NADH- (or NADPH) -dependent oxidoreductase (see Formulas 1 and 2).
Glycerol → 3-HP + H ₂ O (Formula 1)
3-HP + NADH + H ⁺ → 1,3-propanediol + NAD ⁺ (Formula 2)

  1,3-propanediol is not further metabolized, and as a result, accumulates at a high concentration in the culture medium.

  Glycerol is typically used as a starting material for the biological production of 1,3-propanediol. However, glucose and other carbohydrates are also suitable substrates for 1,3-propanediol production. Specifically, Laffend et al. (US Pat. Nos. 5,099,069 and 5,037,096) use a single microorganism comprising dehydratase activity to produce 1 from a carbon substrate other than glycerol or dihydroxyacetone (especially glucose). , 3-Propanediol is disclosed. Emptage et al. (US Pat. No. 5,697,097) distinguishes from non-specific catalytic activity (converting 1,3-propanediol oxidoreductase encoded by dhaT) that converts 3-HP to 1,3-propanediol. The significant increase in titer (grams produced per liter) obtained by Payne et al. (US Pat. No. 5,637,097) disclose certain vectors and plasmids useful for biologically producing 1,3-propanediol. Cervin et al. (US Pat. No. 5,637,097) disclose an improved E. coli strain for high yield production of 1,3-propanediol. Patent document 1, patent document 3, patent document 4, and patent document 5 are incorporated herein by reference in their entirety.

Enzymes involved in converting glycerol to 3-HP are mainly coenzyme B _12- dependent glycerol dehydratase (EC 4.2.2.130) and coenzyme B _12- dependent diol dehydratase (EC .4.2.1.28) is a coenzyme B _12- dependent enzyme. These different but related coenzyme B _12- dependent enzymes have been well studied for their molecular and biochemical properties. The genes for coenzyme B _12- dependent dehydratase include, for example, Klebsiella pneumoniae, Citrobacter freundii, Clostridium pasteurianum, and S. cerevisiae. Klebsiella oxytoca) and Lactobacillus collinoides have been identified (Non-patent Document 2, Non-patent Document 3, and Non-patent Document 4).

Although there are a wide variety of gene nomenclature used in the literature, in each case the coenzyme B _12- dependent dehydratase has a large or “α” subunit, a medium or “β” subunit, and a small or “γ” It consists of three subunits of subunits. These subunits are assembled into α ₂ β ₂ γ ₂ structures to form apoenzymes. Coenzyme B ₁₂ (Activity Cofactor Chemical Species) binds to the apoenzyme to form a catalytically active holoenzyme. Coenzyme B ₁₂ is thereby so involved in the radical mechanism by which catalysis occurs, is required for activity of the catalyst.

Biochemically, both coenzyme B _12- dependent glycerol and coenzyme B _12- dependent diol dehydratase are known to undergo mechanism-based suicide inactivation by glycerol and other substrates (non-patent) Document 3, Non-Patent Document 5). Further inactivation occurs through the interaction of the holoenzyme with high concentrations of 1,3-propanediol. Inactivation cobalt coenzyme B ₁₂ cofactor - involves cleavage of carbon (Co-C) bond results in the formation and inactive cobalamin species 5'-deoxyadenosine. Inactive cobalamin species remain closely bound to the dehydratase and dissociation does not occur without intervention of a coenzyme B _12- dependent dehydratase reactivation factor (“dehydratase reactivation factor”). This inactivation can significantly reduce the reaction rate associated with 3-HP formation, thereby indirectly reducing 1,3-propanediol production.

The effect of coenzyme B _12- dependent dehydratase inactivation can be partially overcome. For example, inactivation can be overcome depending on the protein involved in reactivating dehydratase activity. Dehydratase reactivation factors are described in Patent Document 6, Patent Document 7, Non-Patent Document 3, Non-Patent Document 6, and Non-Patent Document 7. Reactivation occurs in a multi-step process. Initially, in an ATP-dependent process, the interaction between the inactivated coenzyme B _12- dependent dehydratase and the dehydratase reactivator causes the release of a tightly bound inactive cobalamin species, Generate. Subsequently dehydratase apoenzyme and combines with coenzyme B _12, catalytically may be re-form an active holoenzyme, (by enzymatic action in a different ATP- dependent process) inactive cobalamin species complement it may be played to the enzyme _{B 12.} However, since both the dehydratase reactivation and the coenzyme B ₁₂ regeneration process requires ATP, relying only on the dehydratase reactivation factor for dehydratase activity restoration is inherently limited. These ATP-dependent processes are notable for the process of converting glycerol to 3-HP, especially when there is a subsequent reaction from 3-HP to 1,3-propanediol and the 1,3-propanediol concentration is high. Energy load.

As an alternative, the amount of coenzyme B ₁₂ added to the medium during 1,3-propanediol production is increased, or vitamin B (which is converted into coenzyme B ₁₂ in vivo) into the culture medium. recruit _12, it is possible to supply additional coenzyme B ₁₂ to microorganisms either. In either case, however, these addition costs may significantly hinder the economics of the process.

Crowe (Croux) et (Patent Document 8), to develop a process for producing 1,3-propanediol using a recombinant microorganism that expresses a coenzyme _{B 12} -independent dehydratase, coenzyme _{B 12} dependent Addressed issues related to sex dehydratase. However the usefulness of this solution may be limited by the ability to function B ₁₂ -independent Dehidorata in certain preferred process conditions (e.g., under aerobic conditions (Non-Patent Document 8)).

In principle, it should be possible to isolate coenzyme B ₁₂ dependent dehydratase inactivation rate from spontaneous microbial strain was lowered. The reduced inactivation rate increases the turnover rate of coenzyme B _12- dependent dehydratase (product mol / holoenzyme mol) in the microbial host, thereby reducing the required amount of dehydratase and coenzyme B ₁₂ . This approach would also reduce the energy needed to maintain the dehydratase and coenzyme B ₁₂ for that level. In practice, however, the diversity of coenzyme B _12- dependent dehydratase has been found to be limited with respect to inactivation rate.

WO96 / 35796 pamphlet US Pat. No. 5,686,276 International Publication No. 01/012833 Pamphlet US Patent Application No. 60 / 374,931 US Patent Application No. 60/416192 International Publication No. 98/21341 Pamphlet US Pat. No. 6,013,494 International Publication No. 01/04324 A1 Pamphlet Zeng et al., Adv. Biochem. Eng. Biotechnol. 74: 239-259 (2002) Toraya T. , Metal binding enzymes involved in amino acid residues and related radicals (Metaloenzymes Involving Amino Acid-Residue and Related Radicals), Sigel H. et al. And Siegel A.M. Ed; Metal Ions in Biological Systems (Metal Ions in Biological Systems); Marcel Dekker: New York, 1994; Volume 30, pp 217-254 Daniel et al., FEMS Microbiology Reviews, 22: 553-566 (1999). Sauvaget et al., FEMS Microbiology Letters, 209: 69-74 (2002). Seifert et al., Eur. J. et al. Biochem. 268: 2369-2378 (2001) Toraya and Mori, J. et al. Biol. Chem. 274: 3372 (1999) Tobimatsu et al., J. MoI. Bacteriol. 181: 1410 (1999) Hartmanis and Stadman, Arch. Biochem. Biophys. 245: 144-52 (1986)

Thus, the problem to be solved is that currently available coenzyme B _12- dependent dehydratase enzymes cannot provide the reaction rates necessary for industrial applications for the production of industrial compounds.

Applicants
(A) contacting a B _12- dependent dehydratase holoenzyme with a mixture comprising glycerol and at least 25 mM 1,3-propanediol;
(B) screening the B _12-dependent dehydratase holoenzyme, comprising to estimate the turnover ratio of B _12-dependent dehydratase, it provides a method of screening for B _12-dependent dehydratase with improved reaction kinetics To do.

The screening step of the method is
(A) growing cells that do not have a source of coenzyme B ₁₂ , dehydratase reactivating factor, or endogenous B _12- dependent dehydratase on a fermentable carbon substrate that does not contain glycerol;
(B) permeabilizing the cells;
(C) adding the permeabilized cell of step (b) with a mixture of coenzyme B ₁₂ , glycerol, and at least 25 mM 1,3-propanediol;
(D) further comprising quantifying the 3-hydroxypropionaldehyde produced in step (c);
Quantification is a measurement selected from the group consisting of T1, T2, and T (600).

The present invention relates to SEQ ID NOs: 40, 44, 48, 52, 56, 60, 64, 68, 72, 76, 80, 84, 88, 92, 96, 100, 104, 108, 112, 116, 120, 124. 128, 132, 136, 140, 143, 147, 151, 155, 159, 163, 167, 171, 175, 179, 182, 186, 189, 192, 195, 198, 201, 204, 208, 212, 215 218, 221,225,229,233,237,241,245,249,253,257,261,265,269,273,277,281,285,289,293,297,301,304,307,310 313, 316, 319, 322, 325, 328, 331, 334, and 337 Further comprising a nucleic acid sequence encoding a B ₁₂ dependent mutant dehydratase. More preferred _{B 12-dependent} mutant dehydratase, SEQ ID NO: 40,44,48,52,56,60,64,68,72,76,80,84,88,92,96,100,104,108, 112, 116, 120, 124, 128, 132, 136, 140, 143, 147, 151, 155, 159, 163, 167, 171, 175, 179, 182, 186, 189, 192, 195, 198, 201, 204, 208, 212, 215, 218, 221, 225, 229, 233, 237, 241, 245, 249, 253, 257, 261, 265, 269, 273, 277, 281, 285, 289, 293, 297, 301, 304, 307, 310, 313, 316, 319, 322, 325, 328, 331, 334 And 337, and is encoded by a nucleic acid sequence selected from the group consisting of 337. B ₁₂ dependent more preferred nucleic acid sequence encoding a mutant dehydratase, SEQ ID NO: 140,179,186,189,192,195,198,201,212,215,218,301,304,307,310,313 316, 319, 322, 325, 328, 331, 334 and 337 comprising an α-β subunit fusion selected from the group consisting of. B most preferred nucleic acid sequence encoding a _12-dependent mutant dehydratase comprise an alpha-beta subunit fusion selected from the group consisting of SEQ ID NO: 313,322, and 328.

  The invention also includes a nucleic acid sequence further comprising a linker sequence between the α and β subunits of the α-β subunit fusion, wherein the linker sequence is selected from the group consisting of SEQ ID NO: 18 and SEQ ID NO: 19.

The present invention also provides
a) contacting a B _12- dependent dehydratase holoenzyme comprising hot spot 1 or hot spot 2 with a causative agent,
b) having an improved reaction rate comprising screening the mutant produced in step A) for improved kcat and / or stability and c) repeating steps a) and b) B ₁₂ comprising a method of creating dependency dehydratase mutants.

Yet another method of the invention for identifying a B _12- dependent dehydratase having an improved reaction rate compared to a standard dehydratase is:
a) contacting the dehydratase holoenzyme with 5-10 mM glycerol and / or 10-300 mM 1,3-propanediol,
b) measured in at least two time points sufficiently separated to estimate k _cat and total enzyme turnover in a high-throughput assay;
c) comprises selecting a B ₁₂ dependent mutants having kinetics which is improved compared to standard dehydratase.

Yet another method of the present invention to identify the B _12-dependent dehydratase having a reaction rate that is improved compared to the standard dehydratase,
a) contacting the dehydratase holoenzyme with 5-50 mM glycerol and> 300 mM 1,3-propanediol,
b) In a high-throughput assay, measured at one time well away from T0 to estimate the total enzyme turnover number, and c) Dependence on B ₁₂ with improved kinetics compared to standard dehydratase Selecting a mutant strain.

Sequence Listing Applicants have rules governing the disclosure of nucleotide and amino acid sequences in patent applications (Appendix I and II of EPO Commissioner Decision published in EPO Circular OJ 12/1992, Addendum No. 2), 37 C.I. F. R. 1.821-1.825 and Appendix A and B (requirements for disclosure of applications containing nucleotide and / or amino acid sequences), World Intellectual Property Organization (WIPO) standard ST. 25 (1998) and 346 sequences consistent with EPO and PCT sequence listing requirements (Rules 5.2 and 49.5 (a-2), and Detailed Operating Instructions 208 and Annex C) were provided. The sequence description is in accordance with the IUPAC-IYUB standard described in Nucleic Acids Research, 13: 3021-3030 (1985) and Biochemical Journal, 219 (2): 345-373 (1984), incorporated herein by reference. Includes the defined one-letter code for nucleotide sequence properties and three-letter code for amino acids.

  SEQ ID NO: 1 is a 12.1 kB EcoRI-SalI fragment containing wild type GDH isolated from Klebsiella pneumoniae ATCC 25955 (Emptage et al., WO 01/012833 A2) . Wild-type GDH is encoded by the α-subunit (bp 7044-8711), β-subunit (bp 8724-9308), and γ-subunit (bp 9311-9736). The amino acid sequences of the GDH α, β, and γ-subunits are provided as SEQ ID NO: 2, SEQ ID NO: 3, and SEQ ID NO: 4, respectively.

  SEQ ID NOs: 5 and 6 encode primers DHA-F1 and DHA-R1, respectively, used for cloning GDH from plasmid pGH20.

  SEQ ID NO: 7 encodes the reverse transcription primer DHA-R2, which is used for cloning the entire α-subunit and part of the β-subunit of GDH.

  SEQ ID NOs: 8-14 include primers pGD20RM-F1, pGD20RM-R1, TB4BF, TB4BR, pGD20RM-F2, pGD20RM-R2, and GD-C used for regional random mutagenesis of the α-subunit. Code each.

  SEQ ID NOs: 15-17 encode degenerate primers pGD20RM-F3, TB4B-R1, and pGD20RM-R4, respectively, used for the preparation of regional randomized mutant libraries.

  SEQ ID NO: 18 encodes a linker between the α- and β-subunits of the fusion protein Sma3002. SEQ ID NO: 19 encodes a linker between the α- and β-subunits of the fusion protein Xba3009.

  SEQ ID NOs: 20-25 are primers 2-F4-F1,2-F4-R1,12-B1-F1,12-B1-R1,16-H5-used for the synthesis of second generation mutant GDH F1 and 16-H5-R1 are encoded respectively.

  SEQ ID NOs: 26-29 are primers 1-E1-F1, 1-E1-R1, 22-G7-F1, and 22-G7-R1 used to create pure fusion mutants 1-E1 and 22-G7 Is coded respectively.

  SEQ ID NOs: 30-39 are primers 7A-C1-F1, 7A-C1-R1, 7C-A5-F1, 7C-A5-R1, 8-C9-F1 used for the synthesis of Sma3002-derived mutants , 8-C9-R1, 9-D7-F1, 9-D7-R1, 10-G6-F1, and 10-G6-R1 respectively.

  Mutant enzymes derived from wild type GDH are assigned the following SEQ ID numbers according to their nucleic acid and amino acid sequences (Table 1).

  SEQ ID NOs: 340-342 encode primers T53-SM, L509-SM, and V224-SM, respectively.

  SEQ ID NOs: 343 and 344 encode the primers GDHM-F1 and GDHM-R1, respectively.

  SEQ ID NOs: 345 and 346 encode the primers GDHM-F2 and GDHM-R, respectively.

Applicants have improved reaction rates for use in industrial applications, in particular for producing 3-hydroxypropionaldehyde and 1,3-propanediol (ie glycerol and 1,3-propane). Providing an artificial coenzyme B _12- dependent dehydratase (providing an increased total turnover number in the presence of diols) solved the stated problem. These dehydratases created by mutagenesis techniques are characterized by having either an increased k _cat and / or a reduced enzyme inactivation rate compared to the wild-type enzyme produced therefrom. .

The artificial coenzyme B _12- dependent dehydratase of the present invention reduces the inactivation rate without undue sacrifice of the rate of catalysis. This solution to the dehydratase inactivation problem is preferred because it increases the turnover rate of coenzyme B _12- dependent dehydratase (product mol / holoenzyme mol) in the microbial host. The effect of increased turnover reduces dehydratase demand, coenzyme B ₁₂ demand, and energy expenditure to maintain the dehydratase and coenzyme B ₁₂ levels. Furthermore, the coenzyme B _12- dependent dehydratase has been demonstrated to work efficiently for the production of 1,3-propanediol under industrial process conditions.

In addition to providing a series of mutant dehydratases, the present invention also provides two high-throughput assays to facilitate screening of mutant dehydratases. Both methods coenzyme B _12, dehydratase reactivation factor, or in cells that do not have a source of B _12-dependent dependent dehydratase, it is dependent on the presence of B _12-dependent dehydratase apoenzyme form. Therefore, based on the addition of coenzyme B ₁₂ and substrate glycerol, it is possible to accurately control the start of the dehydratase activity.

Specifically, a large library of mutant coenzyme B _12- dependent dehydratase genes is created to express the gene product and a high throughput mode for reduced inactivation in the presence of a mixture of glycerol and 1,3-propanediol Screened. The screening method allows easy and rapid identification of these artificial coenzyme B _12- dependent dehydratases with reduced inactivation rate and improved glycerol catalysis rate. Subsequent dehydratase engineering steps allow the generation and identification of even further improved dehydratase variants.

Based on the teachings herein, as will be described in the present invention, a variety of genes encoding catalysts may B _12-dependent dehydratase activity the conversion of glycerol to 3-HP is, as a target for mutagenesis and screening Appropriateness will be apparent to those skilled in the art. Thus, it is expected that a variety of mutant dehydratases with improved kinetics (ie increased total turnover in the presence of glycerol and 1,3-propanediol) can be identified by means of this invention.

Definitions The following abbreviations and definitions are used for the interpretation of the specification and the claims.

  “Polymerase chain reaction” is omitted in the form of PCR.

  “3-Hydroxypropionaldehyde” is omitted in the form of 3-HP.

The term “dehydratase” is used to refer to any enzyme capable of isomerizing or converting a glycerol molecule to the product 3-hydroxypropionaldehyde. Some dehydratase as described above, require coenzyme B ₁₂ as a cofactor. The terms “coenzyme B _12- dependent dehydratase” and “B _12- dependent dehydratase” are used interchangeably and refer to a dehydratase that requires coenzyme B ₁₂ . The coenzyme B _12- dependent dehydratase is a coenzyme B _12- dependent dehydratase (EC 4.2.1.30) and a coenzyme B _12- dependent dehydratase (EC 4.2.1.28). Comprising. Alternatively, dehydratase may be coenzyme B _{12-independent.} These enzymes do not require coenzyme B ₁₂ as a cofactor, called by the term "B ₁₂ -independent dehydratase". For the purposes of the present invention, the term “glycerol dehydratase” is used to refer specifically to the coenzyme B _12- dependent dehydratase (EC 4.2.1.30) and the term “diol dehydratase”. Is used to refer specifically to the coenzyme B _12- dependent dehydratase (EC 4.2.2.1.28) (the intentional choice of Applicants' nomenclature is to identify the B ₁₂ -independent dehydratase Hartmanis and Stadman, supra, and Croux et al., Supra, referred to as “diol dehydratase” and “glycerol dehydratase”, respectively, should not be confused with the selection.

The term “apoenzyme” refers to the part of an enzyme composed of proteins. Since apoenzymes do not contain non-protein structures that may be required for the enzyme to function, the apoenzymes may be catalytically inactive. The term “cofactor” refers to the non-protein structure required by the apoenzyme for catalytic activity. The term “holoenzyme” refers to a catalytically active protein-cofactor complex. Coenzyme B _12- dependent dehydratase apoenzyme requires cofactor coenzyme B ₁₂ to form a holoenzyme.

The terms “coenzyme B ₁₂ ” and “adenosylcobalamin” are used interchangeably and mean 5′-deoxyadenosylcobalamin.

The terms “vitamin B ₁₂ ” and “cyanocobalamin” are used interchangeably to refer to a derivative of coenzyme B ₁₂ in which the upper axis 5′-deoxy-5′-adenosyl ligand is replaced by a cyano moiety.

“Hydroxocobalamin” refers to a derivative of coenzyme B _{12 in which} the upper shaft 5′-deoxyadenosyl ligand is replaced by a hydroxy moiety. Aquacobalamin is a protonated form of hydroxocobalamin.

Inactivation of the B ₁₂ -independent dehydratase holoenzyme with glycerol or 1,3-propanediol results in the formation of an inactive cobalamin species that has lost the upper axis 5'-deoxy-5'-adenosyl ligand. The term “coenzyme B ₁₂ precursor” refers to a derivative of coenzyme B ₁₂ in which the upper axis 5′-deoxyadenosyl ligand is replaced.

  As used herein, the term “GDH” is specifically isolated from Klebsiella pneumoniae ATCC 25955 and is encoded by bp 7044-8711, bp 8724-9308, and bp 9311-9736 of SEQ ID NO: 1. Refers to B12-dependent glycerol dehydratase. The term “GDH” is used to refer to an assembled complex of α, β, and γ-subunits, and may refer to an apoenzyme or holoenzyme. References to individual subunits of GDH identify subunits such as “GDH α-subunit” or “GDH α-subunit”. Similarly, the α- and β- and γ-subunits may be referred to generically, or to the individual subunits of GDH, and the amino acid sequence of GDH may be referred to. The amino acid sequences of the GDH α, β, and γ-subunits are provided as SEQ ID NO: 2, SEQ ID NO: 3, and SEQ ID NO: 4, respectively.

For purposes of this disclosure, GDH is used as a standard for DNA and amino acid sequences in comparison to artificial (mutant) derivatives. GDH is also used as a standard for wild-type kinetics, against which the kinetics of artificial derivatives created using the present invention are measured. Although GDH is used here as a standard material, any naturally occurring coenzyme B ₁₂ -dependent dehydratase (eg those listed in Table 2) can be used in the present invention indistinguishably from GDH, against which artificial derivatives The reaction rate of is measured.

  The term “genetically altered” refers to the process of changing genetic material by transformation or mutation.

  As used herein, the term “mutant strain” refers to a bacterial clone, plasmid, library or vector containing a GDH enzyme or GDH sequence produced by a mutation process. Alternatively, the term mutant refers directly to the GDH enzyme, GDH amino acid sequence, or GDH DNA sequence produced by the mutation process. Mutant GDH is therefore different from (wild-type) GDH.

The term “improved reaction rate” refers to a reduced dehydratase inactivation rate relative to the wild-type enzyme in the presence of glycerol and / or 1,3-propanediol. Thus, the improved reaction rate is related to the increased total enzyme turnover in the presence of glycerol and 1,3-propanediol. This can be achieved either by increasing k _cat and / or decreasing the rate of enzyme inactivation.

The term “catalytic efficiency” is defined as the enzyme's k _cat / K _M. “Catalytic efficiency” is used to quantify the specificity of an enzyme for a substrate.

The terms “k _cat ”, “K _M ”, and “K _i ” are known to those skilled in the art, (Ferst, Enzyme Structure and Mechanism, 2nd edition. , WH Freeman, New York, 1985, pp 98-120).

The term “k _cat ” is often referred to as “turnover number”. The term “k _cat ” is defined as the maximum number of substrate molecules converted to product per active site per unit time, or the number of rotations of the enzyme per unit time. k _cat = V _max / [E], where [E] is the enzyme concentration (Ferst, Enzyme Structure and Mechanism, 2nd edition, W.H. Freeman, New York, 1985, pp 98-120). The terms “total turnover” and “total turnover number” refer here to the B12-dependent dehydratase holoenzyme with glycerol (T ₀ ) and complete inactivation of the holoenzyme Used to refer to the amount of product formed by the reaction (optionally in the presence of 1,3-propanediol).

The term “K _i ” refers to the inhibition constant of 1,3-propanediol.

The term “k _{observed inactivity} ” refers to the first order kinetics of inactivation observed in the reaction of B _12- dependent dehydratase holoenzyme with excess glycerol and / or excess 1,3-propanediol. Points to a constant. In the presence of excess glycerol alone, the _{observed inactivity} is equal to k _{inactive glycerol} . In the presence of excess 1,3-propanediol only, the _{observed inactivity} is equal to k _{-inactive 1,3-propanediol} . In the presence of both excess glycerol and excess 1,3-propanediol alone, the _{observed inactivity} is equal to a function of both k _{-inactive glycerol} and k _{-inactive 1,3-propanediol} .

The term “T1” refers to the amount of product made by the GDH enzyme reaction measured 30 seconds after the start of the reaction in the presence of 10 mM glycerol and 50 mM 1,3-propanediol. T1 is proportional to k _cat .

The term “T2” refers to the amount of product made by the GDH enzyme reaction measured 40 minutes after the start of the reaction in the presence of 10 mM glycerol and 50 mM 1,3-propanediol. T2 is proportional to k _cat / k _{observed inactivity} and reflects the total enzyme turnover number.

The term “T2 / T1 ratio” refers to the ratio of T2 to T1. The T2 / T1 ratio is proportional to the 1 / k _{observed inactivity} .

The term “T2 (600)” refers to the amount of product made by the GDH enzyme reaction measured 40 minutes after the start of the reaction in the presence of 10 mM glycerol and 600 mM 1,3-propanediol. T2 (600) is proportional to k _cat / k _{observed inactivity} and reflects the total enzyme turnover in the presence of 600 mM 1,3-propanediol.

The term “T (600)” refers to the amount of product made by the GDH enzyme reaction measured 70 minutes after the start of the reaction in the presence of 10 mM glycerol and 600 mM 1,3-propanediol. T (600) is proportional to k _cat / k _{observed inactivity} . The T (600) value gives a more accurate total enzyme turnover than T2 (600) in the presence of 600 mM 1,3-propanediol.

  The terms “polypeptide” and “protein” are used interchangeably.

  The term “host cell” or “host organism” refers to a microorganism that can accept foreign or heterologous genes to express these genes and produce active gene products.

  The term “isolated” refers to a protein or DNA sequence that is removed from at least one component with which it is naturally associated.

  An “isolated nucleic acid molecule” is a polymer of RNA or DNA that is single-stranded or double-stranded, optionally containing synthetic, non-natural or modified nucleotide bases. An isolated nucleic acid molecule in the form of a DNA polymer may comprise one or more fragments of cDNA, genomic DNA or synthetic DNA.

  “Gene” refers to a nucleic acid fragment that expresses a particular protein and may optionally include regulatory sequences preceding (5 ′ non-coding sequences) and following (3 ′ non-coding sequences) the coding sequence. “Native gene” and “wild type gene” refer to a gene found in nature with its own regulatory sequences. “Chimeric gene” refers to any gene that is not a native gene, comprising regulatory and coding sequences that are not found together in nature. Thus, a chimeric gene comprises regulatory and coding sequences that are derived from different sources, or derived from the same source, but arranged in a manner different from that found in nature. It may be. “Endogenous gene” refers to a natural gene in its natural location in the genome of an organism. A “foreign” gene refers to a gene that is not normally found in the host organism, but is instead introduced into the host organism by gene transfer. The exogenous gene can comprise a natural gene inserted into a non-natural organism or a chimeric gene. A “transgene” is a gene that is introduced into the genome by transformation.

  The terms “encoding” and “coding” refer to the process by which a gene generates an amino acid sequence through transcriptional and translational mechanisms. The process of encoding a particular amino acid sequence includes a DNA sequence that may involve base substitutions that do not cause a change in the encoded amino acid, or may alter one or more amino acids, but is encoded by the DNA sequence It is understood to include DNA sequences with base substitutions that do not affect the sensory properties of the protein being processed. Accordingly, the present invention is understood to encompass more than a particular exemplary sequence.

  The term “amino acid” refers to the basic chemical structural unit of a protein or polypeptide. Amino acids are defined according to the IUPAC-IYUB standard described in Nucleic Acids Research, 13: 3021-3030 (1985) and Biochemical Journal, 219 (2): 345-373 (1984), incorporated herein by reference. As identified by either the one letter code or the three letter code for amino acids.

For certain proteins, point substitution mutations within the DNA coding region, and resulting amino acid changes, are identified with reference to standard DNA and amino acid sequences using one of the following notations. For example, in the case of a mutation in GDH, the mutation is described using one of the notations described below.
1. Detailed notation: first the nucleotide sequence of the wild-type codon is presented, followed by the “α”, “β” or “γ” symbol (which distinguishes the α-, β-, or γ-subunit variation of GDH) Therefore, the specific wild type amino acid of the three letter abbreviation and its position are presented. The wild type information is then followed by the specific nucleotide and amino acid modifications present in the referenced mutation. An example of this notation is GGG (α-Gly63) to GGA (Gly), where the 63rd codon of the α-subunit is changed in nucleotide sequence from “GGG” to “GGA” in the mutant. Suffer from silent mutations.
2. “Abbreviated” notation: the symbol “α”, “β”, or “γ” (to distinguish it from a mutation in the α-, β-, or γ-subunit of GDH) followed by a one-letter abbreviation of a wild-type amino acid, It is followed by the codon position and the one letter abbreviation of the mutant amino acid. An example of this notation is α-V224L, which represents a mutation from wild type valine to leucine at codon 224 in the α-subunit in the mutant. It is well known in the art that changes in genes that result in the production of chemically equivalent amino acids at a particular site (but do not affect the encoded protein sensory properties) are common.

  “Substantially similar” refers to a nucleic acid molecule in which one or more nucleotide base changes result in a substitution of one or more amino acids but do not affect the sensory properties of the protein encoded by the DNA sequence. Point to. “Substantially similar” also refers to nucleic acid molecules in which one or more nucleotide base changes do not affect the ability of the nucleic acid molecule to mediate altered gene expression by antisense or co-suppression techniques. “Substantially similar” also refers to substantial changes in the sensory properties of the resulting transcript, or the sensory properties of the resulting protein molecule, to the ability to mediate changes in gene expression by antisense or cosuppression techniques. Refers to modifications (such as deletion or insertion of one or more nucleotide bases) of a nucleic acid molecule of the invention that do not affect the target. The present invention encompasses more than certain exemplary sequences.

  Each proposed modification is well within the routine skill of the art, as is the determination of maintaining the biological activity of the encoded product. Furthermore, those skilled in the art will recognize that similar sequences substantially encompassed by the present invention may be obtained under stringent conditions (0.1 × SSC, 0.1% SDS, 65 ° C. and 2 × SSC, 0.1% SDS). Recognize that they are also defined by their ability to hybridize to the sequences exemplified herein (under washing followed by 0.1 × SSC, 0.1% SDS). Preferred substantially similar nucleic acid fragments of the present invention are nucleic acid fragments whose DNA sequence is at least 80% identical to the DNA sequence of the nucleic acid fragments reported herein. More preferred nucleic acid fragments are at least 90% identical to the DNA sequence of the nucleic acid fragments reported herein. Most preferred are nucleic acid fragments that are at least 95% identical to the DNA sequence of the nucleic acid fragments reported herein.

  A nucleic acid fragment can be “hybridized” with another nucleic acid fragment, such as cDNA, genomic DNA, or RNA, if a single-stranded form of the nucleic acid fragment can anneal with other nucleic acid fragments under appropriate conditions of temperature and solution ionic strength. It is possible. Hybridization and washing conditions are well known, and Sambrook J. et al. Fritsch E .; F. And Maniatis T. et al. Molecular cloning: laboratory manual (Molecular Cloning: A Laboratory Manual), 2nd edition, Cold Spring Harbor Laboratory: Cold Spring Harbor, New York (NY) (1989) in particular. Illustrated in Chapter 11 and Table 11.1 thereof.

  “Substantial part” means by manual evaluation of sequences by those skilled in the art or by BLAST (“Basic Local Alignment Search Tool”, Altschul et al., J. Mol. Biol. 215: 403-410 (1993), see also www.ncbi.nlm.nih.gov/BLAST/) and putative estimation of genes or polypeptides by computer-assisted sequence comparison and identification Refers to an amino acid or nucleotide sequence comprising the amino acid sequence of the polypeptide or gene nucleotide sequence sufficient to obtain an identity. In order to identify a putative polypeptide or nucleic acid sequence as homologous to a known protein or gene, typically 10 or more contiguous amino acids or 30 or more nucleotide sequences are required. In addition, with respect to nucleotide sequences, gene-specific oligonucleotide probes comprising 20-30 contiguous nucleotides are sequence-dependent gene identification methods (eg, Southern hybridization) and isolated (eg, in situ in bacterial colonies or bacteriophage plaques) (In situ) hybridization). Furthermore, in order to obtain a specific nucleic acid molecule comprising a primer, a short oligonucleotide of 12 to 15 bases may be used as an amplification primer in PCR. Thus, a “substantial portion” of the nucleotide sequence comprises sufficient sequence to allow specific identification and / or isolation of the nucleic acid molecule comprising the sequence. The present specification teaches partial or complete amino acid and nucleotide sequences that encode one or more specific proteins. Those skilled in the art can benefit from the sequences reported herein and use all or a substantial portion of the disclosed sequences for purposes known to those of skill in the art. The present invention thus comprises the complete sequences as reported in the attached sequence listing, as well as a substantial portion of these sequences as defined above.

  The term “complementary” describes the relationship between double stranded DNA. For example, for DNA, adenosine is complementary to thymine and cytosine is complementary to guanine. Accordingly, the present invention includes the complete sequences as reported in the accompanying sequence listing, as well as isolated nucleic acid molecules complementary to substantially similar nucleic acid sequences.

  The term “percent identity” is a relationship between two or more polypeptide sequences or two or more polynucleotide sequences determined by comparing sequences, as is known in the art. “Identity” in the art also means the degree of sequence relatedness between polypeptide or polynucleotide sequences, as the case may be, as determined by alignment between strings of such sequences. “Identity” and “Similarity” are: ) "Computational Molecular Biology", Lesk A. et al. M.M. Ed., Oxford University, New York, 1988,2. ) “Biocomputing: Informatics and Genome Projects”, Smith D .; W. Ed., Academic, New York, 1993,3. ) "Computer Analysis of Sequence Data", Part 1, Griffin A. et al. M.M. , And Griffin H. et al. G. Hen, Humana, New Jersey, 1994,4. ) "Sequence Analysis in Molecular Biology", von Heinje G. et al. Ed., Academic, New York, 1987,5. ) “Sequence Analysis Primer”, Gribskov M .; And Devereux J. et al. Can be readily calculated by known methods, including but not limited to those described in Hen, Stockton, New York, 1991. Preferred methods for determining identity are designed to give the greatest match between the sequences tested.

  The method of determining identity and similarity is organized with commonly available computer programs. Preferred computer program methods for determining identity and similarity between two sequences include Needleman and Wunsch whose standard default values are gap formation penalty = 12 and gap extension penalty = 4. ) GCG pile-up program (Devereux et al., Nucleic Acids Res. 12: 387-395 (1984)), BLASTP, BLASTN, and FASTA, (Pearson et al., Proc) in the GCG program package using algorithms Natl.Acad.Sci.USA, 85: 2444- 2448 (1988)), but is not limited thereto. The BLASTX program is generally available from NCBI and other sources ("BLAST Manual", Altschul et al., Natl. Cent. Biotechnol. Inf., Natl. Library Med. (NCBINLM) NIH, Bethes. MD; Altschul et al., J. Mol. Biol., 215: 403-410 (1990); Altschul et al., Nucleic Acids Res., 25: 3389-3402 (1997)). Another preferred method of determining identity is by the DNASTAR protein alignment protocol method using the Jotun-Hein algorithm (Hein et al., Methods Enzymol., 183: 626-645 (1990)). The default variables for the Jotun-Hein method for alignment are: ) For multiple alignments, gap penalty = 11, gap length penalty = 3, and 2. ) In paired alignment, ktuple = 6. By way of example, a polynucleotide having a nucleotide sequence that has “identity” of at least 95%, for example, relative to a reference nucleotide sequence means that the polynucleotide sequence includes 5 mutations for every 100 nucleotides of the reference nucleotide sequence. Except as good, it means that the nucleotide sequence of the polynucleotide is identical to the reference sequence. In other words, to obtain a polynucleotide having a nucleotide sequence that is at least 95% identical to the reference nucleotide sequence, up to 5% of the nucleotides in the reference sequence may be deleted or replaced with another nucleotide; Alternatively, the number of nucleotides up to 5% of all nucleotides in the reference sequence may be inserted into the reference sequence. These variations of the reference sequence may occur at the 5 ′ or 3 ′ terminal site of the reference nucleotide sequence, or at any position between the terminal sites, in the reference sequence or in one or more of the reference sequences It may occur by being individually inserted into nucleotides in adjacent groups. Similarly, a polypeptide having an amino acid sequence that has at least 95% identity to a reference amino acid sequence, for example, includes up to 5 amino acid changes for each 100 amino acids of the reference amino acid. Otherwise means that the amino acid sequence of the polypeptide is identical to the reference sequence. In other words, in order to obtain a polypeptide having an amino acid sequence that is at least 95% identical to the reference amino acid sequence, up to 5% of the amino acid residues in the reference sequence may be deleted or substituted with another amino acid, Alternatively, the number of amino acids up to 5% of all amino acid residues in the reference sequence may be inserted into the reference sequence. These changes in the reference sequence may occur at the amino or carboxy terminal site of the reference amino acid sequence, or one or more contiguous groups in the reference sequence or at any position between these terminal sites It may occur by being inserted individually between the residues.

  The term “homologous” refers to a protein or polypeptide that is native or naturally occurring in a particular host cell. The present invention includes microorganisms that produce homologous proteins through recombinant DNA technology.

  The term “percent homology” refers to the degree of amino acid sequence identity between polypeptides. When the first amino acid sequence is identical to the second amino acid sequence, the first and second amino acid sequences exhibit 100% homology. Homology between any two polypeptides is a linear function of the total number of matched amino acids at a particular site in either sequence, for example, if half of the total number of amino acids in either sequence is identical, The sequence is said to show 50% homology.

  “Codon degeneracy” refers to a branch in the genetic code that allows for variations in the nucleotide sequence that do not affect the amino acid sequence of the encoded polypeptide. Those skilled in the art are familiar with the “codon-bias” exhibited by a particular host cell in the use of nucleotide codons to specify a particular amino acid.

  Sequence modifications such as deletions, insertions or substitutions in the sequence that produced a silent change that did not substantially affect the functional properties of the resulting protein molecule were also considered. For example, changes in the gene sequence that reflected the degeneracy of the genetic code or resulted in the production of chemically equivalent amino acids at specific sites were considered. In some cases it may actually be desirable to create sequence variants to study the effects of changes in the biological activity of a protein. Each proposed change is well within the routine skill of the art, as is the determination of biological activity retention in the encoded product.

  The term “expression” refers to the transcription and translation of a gene product sequence from the genetic code to the gene product.

  The terms “plasmid”, “vector”, and “cassette” often carry genes that are not part of the central metabolism of a cell and refer to extrachromosomal elements that are usually in the form of circular double-stranded DNA molecules. Point to. Such elements may autonomously replicate single or double stranded DNA or RNA sequences, genomic integration sequences, linear or circular phage or nucleotide sequences derived from any source, where Several nucleotide sequences are ligated or recombined into a unique configuration, which can introduce promoter fragments and DNA sequences into the cell, along with the appropriate 3 'untranslated sequences for the selected gene product. “Transformation cassette” refers to a specific vector containing a foreign gene and having elements in addition to the foreign gene that facilitate transformation of a particular host cell. “Expression cassette” refers to a specific vector containing a foreign gene and having elements in addition to the foreign gene that allow for enhanced expression of that gene in a foreign host.

  The term “genetic recombination” refers to a process by which a genetic combination that does not exist in a parent template molecule is formed by a process of crossover or independent combination. Thus, genetic recombination includes all combinations of genetic sequences that can be obtained from a parent template molecule (where each nucleotide position of a newly generated “GMO product” is the parent of any particular nucleotide position. Furthermore, genetic recombination involves the introduction of new mutations (ie deletions, substitutions, insertions).

  The term “genetically modified polypeptide” means a polypeptide encoded by a genetically modified gene or DNA. Genetically modified polypeptides often have altered or enhanced properties.

  The term “altered property” as applied to a polypeptide or protein refers to a property associated with a protein encoded by a nucleotide sequence that can be measured by an assay, and that property is associated with a native sequence. Increased or decreased compared. Examples of preferred properties of enzymes that may be modified include enzyme activity, substrate specificity, stability to inhibitors, thermal stability, protease stability, solvent stability, detergent stability, and folding properties. It is done. “Enhanced biological property” refers to an altered property over that associated with the native sequence. A “reduced biological property” refers to an altered property that is less than that associated with the native sequence.

  The term “template” or “parent template” refers to a nucleic acid molecule that is copied by a DNA or RNA polymerase to produce a new DNA or RNA strand according to the rules of Watson-Crick base pairing. Since the first copy generated from the template molecule has a complementary sequence, the sequence information in the template (or “model”) is preserved. The template molecule may be single or double stranded and may be derived from any source.

  “Replication” is the process by which a complementary copy of a nucleic acid strand of a “template nucleic acid” is synthesized by a polymerase enzyme. In “primer-induced” replication, this process requires a hydroxyl group (OH) at the 3 ′ position of the (deoxy) ribose moiety of the terminal nucleotide of the “double” “oligonucleotide” to initiate replication. .

  The “5 ′ region” and “3 ′ region” of a nucleic acid are used as relative terms with reference to a nucleotide region in which genetic recombination is desired to occur. These regions may be in the template molecule or in the lateral DNA sequence attached to the template molecule. Unpaired primers anneal to portions of these 5 'and 3' regions.

  A “lateral sequence” or “lateral DNA fragment” refers to the 5 ′ or 3 ′ of a template molecule to provide a unique nucleotide sequence (relative to the template molecule) that an unpaired primer may anneal to. Refers to a short piece of DNA attached to any of the regions.

  A “full-length extension product” is a nucleotide produced by primer-induced replication that has a length very similar to that contained between the 5 ′ and 3 ′ regions of the parent template (within about 100 bases). Is an array.

  “Amplification” is a process in which replication is repeated in a cyclic fashion such that copies of several “template nucleic acids” are increased in either a linear or logarithmic manner.

  The term “primer” refers to an oligonucleotide (synthetic or natural) that can act as a starting point for nucleic acid synthesis or replication along a complementary strand when subjected to conditions in which the synthesis of the complementary strand is catalyzed by a polymerase. The primer must have the ability to anneal to the complementary strand based on sequence complementarity between the primer itself and the complementary strand of the nucleic acid, but some base mismatch is allowed. Primer size, base sequence, complementarity, and target interaction requirements are discussed in more detail below. The term “primer” itself is generally used herein by the applicant and encompasses any sequence-linked oligonucleotide that functions to initiate the nucleic acid replication process (ie, can prime synthesis).

  The term “forward primer” refers to a primer that can prime synthesis at the 5 ′ region of the sense strand of a double-stranded template molecule or prime synthesis at the 5 ′ region of a single-stranded template molecule.

  The term “reverse transcription primer” refers to the 5 ′ region of a single-stranded template molecule that can prime synthesis in the 5 ′ region of the double-stranded template molecule antisense strand, or the antisense strand of the double-stranded template molecule. Refers to a primer that can prime synthesis.

  The term “pairing primer” consists of forward and reverse transcription primers that are designed to anneal to a single template molecule and allow the synthesis of an exact copy of the template through a primer-induced nucleic acid amplification process. Refers to a primer pair. For double-stranded template molecules, the forward primer produces an exact copy of the antisense strand (complementary copy of the sense strand used as the template) and the reverse transcription primer is the sense strand (antisense used as the template) The forward and reverse transcription primers allow the synthesis of an exact copy of the double stranded template, as it produces an exact copy of the complementary copy of the strand). In contrast, if the template molecule is single stranded, an exact copy of the template is generated using a primer-induced nucleic acid amplification process.

  The term “unpaired primer” refers to forward and reverse transcription primers that are not designed to anneal to a single template molecule and allow the synthesis of an exact copy of that template through a primer-induced nucleic acid amplification process. Refers to a pair of primers consisting of Instead, the forward primer anneals to the first template molecule but cannot anneal to the second template molecule. The reverse transcription primer anneals to a second template molecule that differs in sequence from the first template molecule, but cannot anneal to the first template molecule. This unique unpaired primer design ensures that single-stranded or double-stranded template molecules cannot be amplified by the primer-induced nucleic acid amplification process unless genetic recombination occurs during replication through template switching.

  The term “primer-induced extension” refers to any method known in the art in which primers are used to aid in the replication of nucleic acid sequences in linear or logarithmic amplification of nucleic acid molecules. For example, primer-induced extension is known in the art, including but not limited to polymerase chain reaction (PCR), ligase chain reaction (LCR), and strand displacement amplification (SDA). It may be performed by any of several schemes.

For conversion of GDH and encoding genes glycerol to 3-HP, dehydratase responsible for catalyzing this reaction may be a B ₁₂ dependent dehydratase or B _{12-independent} dehydratase. However, the purpose of the present invention, dehydratase is B ₁₂ dependent dehydratase. The _{B 12-dependent} dehydratase can be GDH, a glycerol dehydratase (E.C.4.2.1.30), or a diol dehydratase (E.C.4.2.1.28). Each of these enzymes has an α ₂ β ₂ γ ₂ structure. There are a wide variety of gene nomenclatures used in the literature, and for clarity, Klebsiella pneumoniae, Citrobacter freundii, Clostridium pasteurianum, and Rat typhimurium. The gene name and GenBank number of the dehydratase gene of the fungus (Salmonella typhimurium), Lactobacillus collinoide, and Klebsiella oxytoca are shown in Table 2 and are easily identified. As for genes, for example, Daniel et al. (FEMS Microbiol. Rev., 22: 553, (1999)) and Toraya and Mori (J. Biol. Chem., 274: 3372) ( 1999)).

For the purposes of the present invention, encoded by dhaB1, dhaB2, and dhaB3 (SEQ ID NO: 1 and SEQ ID NO: 2, 3, and 4 respectively) of K. pneumoniae (WO 01/012833 A2) GDH is used as a target for mutagenesis. This enzyme, glycerol dehydratase, is known to have a preferred substrate of glycerol, consisting of two 63 kDa α subunits, two 21 kDa β subunits, and two 16 kDa γ subunits. . However, those skilled in the art will be aware of the techniques described herein, such as Citrobacter freundii, Clostridium pasteurianum, or other strains of K. pneumoniae strain glycerol dehydratase, or Salmonella s. A diol dehydratase from Typhimurium), Klebsiella oxytoca or K. pneumoniae will also be recognized as appropriate. Similarly any gene whose activity encoding catalysts may B _12-dependent dehydratase activity conversion to 3-HP of glycerol, amino acid substitutions that do not change the functionality of GDH enzyme, any amino acid sequence including deletion or addition It should be suitable as a target for the present invention. Thus, those skilled in the art will appreciate that genes encoding GDH isolated from other sources are also suitable for use in the present invention.

B _12-dependent dehydratase inactivation B _12-dependent dehydratase in glycerol or 1,3-propanediol presence reversibly inactivated by the interaction of the glycerol in the catalysis, or with 1,3-propanediol Suffer. Inactivation cobalt coenzyme B ₁₂ cofactor - involves cleavage of carbon (Co-C) bond results in the formation of 5'-deoxyadenosine and an inactive cobalamin species. Inactive cobalamin species keeps a close coupling of the dehydratase, without intervention of coenzyme B ₁₂ dependent dehydratase reactivating factor does not occur dissociation.

FIG. 1 shows a typical time course associated with GDH inactivation. These enzyme activity assays are encoded by dhaB1, dhaB2, and dhaB3 (SEQ ID NO: 1 and SEQ ID NO: 2, 3, and 4 respectively) of K. pneumoniae (WO 01/012833 A2). This is implemented using GDH. Figure 1 upper curve, while indicating the time course of the reaction of GDH with 10mM glycerol as a substrate _{(K m} about 0.5 mM), the lower curve, assay of 50 mM 1,3-propanediol (K _i about 15 mM). Clearly, the rate of 3-HP product formation decreases rapidly with time as inactivation occurs according to the first-order inactivation rate constant (k _{observed inactivity} ).

Mutant GDH with improved activity
Based on the observed GDH inactivation, a series of mutant GDHs with reduced inactivation rate compared to wild type GDH was created. Typically approaches involve creating and isolating mutant enzymes with increased total turnover in the presence of glycerol and 1,3-propanediol. This can be achieved either by increasing k _cat and / or decreasing the rate of enzyme inactivation.

  The process of improving GDH activity involves the construction of an expression vector comprising the GDH gene, mutagenesis of the GDH coding sequence, and finally the isolation of mutants with a reduced inactivation rate. Subsequent mutagenesis steps allow the development of GDH coding sequences.

A mutant B _12- dependent dehydratase library can be prepared using any wild-type (or substantially similar) B _12- dependent dehydratase as a starting material for mutagenesis.

B ₁₂ For traditional methods B _12-dependent dehydratase mutagenesis mutagenesis dependent dehydratase may be used a variety of approaches. Two suitable approaches used here include error-prone PCR (Leung et al., Techniques, 1: 11-5 (1989); Zhou et al., Nucleic Acids Res., 19: 6052-6052. (1991); and Speed et al., Nucleic Acids Res., 21: 777-778 (1993)), and in vivo mutagenesis.

A major advantage of error-prone PCR is that all mutations introduced by this method is B ₁₂ dependent within dehydratase gene is that any changed by changing the PCR conditions can be easily controlled. Alternatives include the E. coli XL1-Red strain and the Epicurian coli XL1-Red mutator strain (Greener (Stratagene, La Jolla, Calif.)) From Stratagene, La Jolla, Calif. In vivo mutagenesis may be used using commercially available materials such as Greener and Callahan, Strategies, 7: 32-34 (1994)). In this strain, three of the primary DNA repair pathways (mutS, mutD, and mutT) are defective, resulting in a 5000-fold higher mutation rate than the wild type. In vivo mutagenesis does not depend on ligation efficiency (like error-prone PCR), but mutations can occur in any region of the vector and the mutation rate is generally much lower.

Alternatives include the “gene shuffling” method (US Pat. Nos. 5,605,793, 5,811,238, 5,830,72, and 5,837). , 458 Pat), or using any similar means of promoting the activity of a gene recombination occurs between the nucleic acid may be constructed a mutant strain B _12-dependent dehydratase having a deactivation rate decreased It is considered. Gene shuffling is particularly attractive because of its easy performance and fast mutagenesis. The process of gene shuffling involves limiting the gene of interest to specific size fragments in the presence of an additional population of DNA regions similar (or different) to the gene of interest. The pool of fragments is then denatured and reannealed to create a mutated gene. The mutated gene is then screened for altered activity.

By mutating the wild-type B _12-dependent dehydratase sequences, the activity that is altered or enhanced by this method may be screened. The sequence should be double stranded and can be of various lengths ranging from 50 bp to 10 kB. The sequence may be randomly digested into fragments ranging from about 10 bp to 1000 bp using restriction endonucleases well known in the art (Maniatis, ibid). In addition to the full length sequence, a population of fragments capable of hybridizing to all (or part) of the sequence may be added. Similarly, a population of fragments that are not hybridizable to the wild type sequence may be added. Typically, these additional fragment populations are added in excess of about 10 to 20 times by weight compared to the total nucleic acid. In general, this process allows the production of about 100 to 1000 different specific nucleic acid fragments in a mixture. The mixed population of random nucleic acid fragments is denatured to form single stranded nucleic acid fragments and then reannealed. Only single-stranded nucleic acid fragments having regions of homology with other single-stranded nucleic acid fragments will reanneal. Random nucleic acid fragments may be denatured by heating. One skilled in the art can determine the conditions necessary to completely denature the double-stranded nucleic acid. Preferably, the temperature is about 80 ° C to 100 ° C. The nucleic acid fragment may be reannealed by cooling. Preferably, the temperature is about 20 ° C to 75 ° C. Regeneration can be accelerated by the addition of polyethylene glycol (“PEG”) or salt. The salt concentration is preferably 0 mM to 200 mM. The annealed nucleic acid fragments are then incubated in the presence of nucleic acid polymerase and dNTPs (ie dATP, dCTP, dGTP and dTTP). The nucleic acid polymerase may be Klenow fragment, Taq polymerase or any other DNA polymerase known in the art. A polymerase may be added to the random nucleic acid fragments prior to annealing, simultaneously with annealing, or after annealing. The cycle of denaturation, regeneration, and incubation in the presence of polymerase is repeated as many times as desired. Preferably the cycle is repeated 2-50 times, more preferably the procedure is repeated 10-40 times. The resulting nucleic acid is a larger double-stranded polynucleotide of about 50 bp to about 100 kB and may be screened for expression and altered activity by standard cloning and expression protocols (Maniatis, ibid.).

In addition to the methods exemplified above (which are designed to directly mutagenesis the gene encoding the B _12-dependent dehydratase), they can be used for the purposes mentioned traditional methods of creating mutants here. For example B ₁₂ wild-type cells having any dependency dehydratase activity by exposure to various agents such as radiation or chemical mutagens and then may be screened for the desired phenotype. When creating mutations by radiation, either ultraviolet (UV) or ionizing radiation may be used. The wavelength of short wavelength UV suitable for genetic variation is in the range of 200 nm to 300 nm, preferably 254 nm. UV radiation within this wavelength mainly causes intra-nucleic acid sequence changes from guanidine and cytosine to adenine and thymidine. All cells have a DNA repair mechanism that repairs most UV-induced mutations, so agents such as caffeine and other inhibitors are added to interfere with the repair process and some effective Mutations may be maximized. Longwave UV mutations using light in the range of 300 nm to 400 nm are also possible, but this range is generally short wavelength UV unless used in combination with various activators that interact with DNA (such as psoralen dyes). Not as effective as light. Similarly, mutagenesis with chemical agents is effective in generating mutant strains, and commonly used materials include chemicals that affect non-replicating DNA (such as HNO ₂ and NH ₂ OH) and replicate. Examples include agents that affect DNA (such as acridine dyes that should be noted by causing a frameshift mutation). Specific methods for creating mutants using radiation or chemical agents are well documented in the art. For example, Thomas D. Biotechnology: A Textbook of Industrial Microbiology, 2nd edition, Sinauer Associates (Sunderland, Mass.), Sunderland, Mass. Deshandande, Mukund) V. Appl. Biochem. Biotechnol. 36: 227 (1992).

Regardless of the mutagenesis method, the gene may be developed so that the enzyme has an increased total turnover number in the presence of glycerol and 1,3-propanediol. This can be achieved either by increasing k _cat and / or decreasing the rate of enzyme inactivation.

The mutagenesis Figure 5 B _12-dependent dehydratase using recombinant generation stretching method using unpaired primers, based on the genetic modification of the two genes, the principle of the recombinant generation stretching method using unpaired primers Show. This method requires that the parent genes have different DNA sequences at their 5 ′ and 3 ′ ends. If the parent gene has identical 5 ′ and 3 ′ sequences, a short lateral DNA fragment has been attached to the 5 ′ or 3 ′ end of the gene by standard PCR (as shown in step A of FIG. 5). Must-have.

  Then, following the design of two unpaired primers for thermal cycling, the PCR product can be used as a template for genetically engineered extension methods. Primer-1 anneals to the 5 'end of template-1 but does not bind to the 5' end of template-2. Primer-2 anneals to the 3 'end of template-2 but does not bind to the 3' end of template-1 (step B). This ensures that neither parent template is amplified by the thermal cycling reaction.

  A short anneal and synthesis cycle is performed to create a series of short DNA fragments (step C). If there is sufficient homology, some of these DNA fragments will anneal to a different template (ie, “template switch”) in a subsequent annealing cycle (step D). Subsequently, a recombinant DNA fragment is made as shown in FIG. Finally, a recombinant DNA gene having a 5 'end of template-1 and a 3' end of template-2 is created (step E).

  At this point, the previous unpaired primer-1 and primer-2 become paired primers for the newly created recombinant gene. A further annealing and synthesis cycle amplifies a pool of recombinant DNA genes having the 5 'end of template-1 and the 3' end of template-2 (step F of FIG. 5). During the amplification step, the recombinant DNA gene can be further genetically modified, thereby increasing the number of crossovers of the recombinant DNA gene. Theoretically, all amplified products should be recombinant DNA products. These recombinant products can be derived directly from the parent template molecule, or additional mutations (eg, insertions, deletions, and substitutions) may be incorporated into the final genetically modified product. There is negligible contamination of the parent template in the final reaction mixture.

  More details on this methodology are disclosed in US patent application Ser. No. 0 / 374,366, incorporated herein by reference.

The B _12-dependent dehydratase variant having a B ₁₂ dependent identified improved speed performance of the dehydratase mutants with improved speed performance (from a large population) in order to identify, develop high-throughput assays, screening Make it very easy. Disclosed herein is a simple screening method that provides an improvement in k _cat compared to a standard (eg, wild type GDH) and / or an estimate of the total turnover number of a dehydratase variant, depending on two measurements. . Specifically, a GDH mutant gene library is cloned into E. coli for GDH expression by standard methods, an isolate is obtained, cultured in Lennox broth, permeabilized, and then Assayed for GDH activity. Although screening methods are described for GDH contained in whole cells, those skilled in the art will recognize that the methods can be easily modified and applied to enzymes in crude or purified preparations.

The GDH assay developed here relies on the presence of the apoenzyme form of the GDH enzyme in cells that do not have a coenzyme B ₁₂ , dehydratase reactivator, or B _12- dependent dehydratase source. GDH holoenzyme catalytically active form only by the addition of coenzyme B ₁₂ into the culture solution. Thus, the reaction of GDH can be switched “on” with high precision and efficiency by the addition of coenzyme B ₁₂ and the substrate glycerol. This allows for precise control of the duration (timing) of GDH activity, thus allowing accurate measurement of GDH activity. The reaction starts at 0:00 (time = t ₀ ) and product formation is measured during the initial phase of the reaction (immediately after t ₀ and before enzyme inactivation occurs, time = t ₁ ). Furthermore, product formation is measured even longer (immediately after t ₀ and after completion of enzyme inactivation, time = t ₂ ). Determination of product (3-HP) formation is based on colorimetric aldehyde analysis (Zurek G., and U. Karst, Analytica Chimica Acta, 351: 247-257 (1997)).

In GDH of coenzyme B ₁₂ and glycerol presence of saturating concentrations, the amount of 3-HP with time generated (y) is estimated by the following.
y = T ₀ + AMP (1-exp (−k _{observed inactivity} ^* time))
In the formula, T ₀ is the amount of 3-HP at t ₀ (background), and AMP is the amount of 3-HP (total metabolism) produced from t ₀ until GDH is completely inactivated. The _{observed inactivity} is the observed first-order inactivation rate constant. In the absence of 1,3-propanediol, the _{observed inactivity} is all due to k _{-inactive glycerol} (first-order inactivation rate constant due to glycerol) (see the upper curve in FIG. 1). . AMP is equal _{to the observed inactivity} [GDH] ^* k _cat / k. The time course is used to establish appropriate times (t ₁ and t ₂ ) for fixed concentrations of GDH, coenzyme B ₁₂ , and glycerol. After correcting for T ₀ , the amount of product formed at t ₁ (T 1, before enzyme inactivation occurs) is used to estimate [GDH] ^* k _cat and the product formed at t ₂ (T 2, enzyme Is used to estimate the total turnover number and [GDH] ^* k _cat / k _{observed inactivity} . Thus, the T1 value is directly proportional to k _cat , the T2 value is directly proportional to k _cat / k _{observed inactivity} , and T2 / T1 gives 1 / k _{observed inactivity} .

After determining the appropriate time (t ₁ and t ₂ ) for a fixed concentration of wild-type GDH, coenzyme B ₁₂ and glycerol (in the absence of 1,3-propanediol), the values of T1 and T2 are determined under the same assay conditions. However, it was measured again with the addition of 1,3-propanediol (lower curve in FIG. 1). Competitive inhibition occurs in the presence of glycerol and 1,3-propanediol, and the _{observed inactivity} is k _gly (first-order inactivation rate constant due to glycerol) and k _{1,3-propanediol} (1,3 -A function of both the first-order inactivation rate constant due to propanediol). At the appropriate 1,3-propanediol concentration, the T1 value decreases measurablely from the value in the absence of 1,3-propanediol (reflecting competitive inhibition of 1,3-propanediol and glycerol) and T2 The value was significantly reduced from the value in the absence of 1,3-propanediol (reflecting the fact that k _{1,3-propanediol} > k _gly ). In a two-point assay, glycerol is present at a concentration of 5-50 mM, preferably 10 mM, and 1,3-propanediol is present at a concentration of 10-300 mM, preferably 50 mM.

Applicants established conditions for examining wild-type GDH-versus mutant strains in the presence of glycerol and 1,3-propanediol. At the microscopic level, the introduction of mutations into GDH may affect the enzyme affinity k _cat for glycerol and / or 1,3-propanediol, and the respective k _inactivity values. Initially there was no guarantee that the affinity and rate constants would vary independently of each other, but variations in these parameters were performed in the presence of 10 mM glycerol and 50 mM 1,3-propanediol (two-point assay). It was useful to consider how the results of T1 and T2) can be affected. Specifically, slow inactivation (k _observed decrease in _inactivity ) results in an increased T2 / T1 ratio. On the other hand, variations in either T1 or T2 can be attributed to changes in the intrinsic properties of the enzyme. For example, if the relative affinity of glycerol and 1,3-propanediol is constant and k _cat and k _inactivity values are reduced proportionally, T2 / T1 increases but T1 decreases. On the other hand, mutants that have normal interaction with glycerol but have a reduced relative affinity for 1,3-propanediol or a lower rate constant for inactivation of 1,3-propanediol are almost normal. It should show T1, elevated T2, and higher T2 / T1 ratio. Mutants with normal glycerol and 1,3-propanediol affinity but with reduced k _inactivity for either compound should also show normal T1 and show increased T2 and T2 / T1 ratios It is. _An increase in k _cat in k _inactivity or no change in substrate or 1,3-propanediol affinity increases T1 and T2, but does not increase the T2 / T1 ratio. Despite the many variations described above, the aim of the study here is to increase the total turnover number of the GDH enzyme in the presence of glycerol and 1,3-propanediol. This can be achieved either by increasing k _cat and / or decreasing the rate of enzyme inactivation. Therefore, compared to the wild type, a mutant exhibiting at least one improvement selected from the group consisting of higher T1 and / or higher T2 and / or higher T2 / T1 values, or any combination of these parameters, Is considered to have improved properties.

If the glycerol concentration is low compared to the 1,3-propanediol concentration, the measurement of T1 (for estimation of k _cat ) is not possible and thus significant inactivation by 1,3-propanediol occurs. These conditions are reasonable because a high 1,3-propanediol concentration and a low glycerol concentration are desirable in the 1,3-propanediol production process. Recognizing this aspect of the problem, as described above, except that glycerol is present at a concentration of xx to yymM, preferably 10 mM and 1,3-propanediol is present at a concentration of greater than 300 mM, preferably 600 mM. Thus, a single point assay is provided. In this one-point assay, the time is indicated by the t _endpoint , the amount of 3-HP produced is indicated by T (XXX), and XXX is the concentration (mM) of 1,3-propanediol.

Identification of Critical Amino Acids that Affect Coenzyme B ₁₂ Inactivation Applicants disclose a variety of mutant GDH enzymes with reduced coenzyme B ₁₂ inactivation rates compared to the wild type gene. These mutants were identified using the mutagenesis and screening methods described above. Mutants, altered amino acid residues, 1 / k _{observed inactivity} (measured as T2 / T1), and T2 are summarized in Table 1. The SEQ ID number of the DNA sequence of the enzyme is listed in the first column of the table.

  Additional mutant strains identified using the mutagenesis and screening methods described above are summarized in Table 4. This table presents information on each mutant, altered amino acid residue, and T (600) activity. The SEQ ID number of the DNA sequence of the enzyme is listed in the first column of the table.

A variety of mutant GDH enzymes with reduced coenzyme B ₁₂ -inactivation rates are disclosed above. Preferred mutants of the present invention include SEQ ID NOs: 40, 44, 48, 52, 56, 60, 64, 68, 72, 76, 80, 84, 88, 92, 96, 100, 104, 108, 112, 116, 120, 124, 128, 132, 136, 140, 143, 147, 151, 155, 159, 163, 167, 171, 175, 179, 182, 186, 189, 192, 195, 198, 201, 204, 208, 212, 215, 218, 221, 225, 229, 233, 237, 241, 245, 249, 253, 257, 261, 265, 269, 273, 277, 281, 285, 289, 293, 297, 301, 304, 307, 310, 313, 316, 319, 322, 325, 328, 331, 334, and 337 It is below. More preferred variants of the present invention are SEQ ID NOs: 140, 179, 186, 189, 192, 195, 198, 201, 212, 215, 218, 301, 304, 307, 310, 313, 316, 319, 322, 325. 328, 331, 334, and 337. The most preferred variants of the present invention are SEQ ID NOs: 313, 322, and 328.

In addition to the above mutants, a large pool of mutants with improved kinetics can be synthesized, each further comprising some combination of the mutations listed in Tables 2 and 3. For example, each mutation in a mutant strain GDH having a multipoint mutation can be evaluated for a desired effect for improving the overall reaction rate of the enzyme. If desired, any mutation that did not increase the reaction rate can be reverted to the wild type sequence. Similarly, the above list of GDH mutants having only a single point mutation can be combined with another mutation disclosed above to further improve the activity of the enzyme. Applicants disclose the following useful mutations in the GDH enzyme, which use a variety of combinations in combination with an alternative mutant GDH that has a reduced rate of coenzyme B ₁₂ inactivation compared to the wild-type gene: Enzymes may be produced. These mutations include:
1. In α-subunit: TCA (α-Ser41) to TCG (Ser), GTG (α-Val44) to GCG (Ala), GTG (α-Val44) to GAG (Glu), GGT (α-Gly47) to GGC (Gly), CAG (α-Gln59) to CGG (Arg), ATG (α-Met62) to GTG (Val), ATG (α-Met62) to ACG (Thr), ATG (α-Met62) to CTG (Leu) ), ATC (α-Ile63) to GTC (Val), GGG (α-Gly63) to GGA (Gly), CGA (α-Arg65) to CAA (Gln), ATC (α-Ile67) to GTC (Val), TAC (α-Tyr70) to AAC (Asn), GTT (α-Val74) to GTC (Val), GTT (α- al74) to ATT (Ile), ACG (α-Thr77) to GCG (Ala), GTG (α-Val86) to GAG (Glu), CAC (α-His96) to CAT (His), ATC (α-Ile102) To ACC (Thr), ATC (α-Ile102) to GTC (Val), ATC (α-Ile105) to ATT (Ile), GTC (α-Val115) to GCC (Ala), GAG (α-Glu116) to GAA (Glu), GCG (α-Ala119) to ACG (Thr), GTG (α-Val124) to GCG (Ala), CGT (α-Arg134) to CGC (Arg), CGG (α-Arg137) to AGG (Arg) ), AAC (α-Asn141) to ATC (Ile), TGC (α-Cys143) To TGT (Cys), CTC (α-Leu148) to CGC (Arg), AAA (α-Lys149) to AGA (Arg), AAA (α-Lys149) to CAA (Gln), GAT (α-Asp150) to CAT (His), GAT (α-Asp150) to GAC (Asp), AAT (α-Asn151) to AAC (Asn), CCG (α-Pro152) to CCC (Pro), TCA (α-Ser168) to CCA (Pro ), TGC (α-Cys193) to AGC (Ser), GAG (α-Glu209) to GAA (Glu), ATG (α-Met214) to TTG (Leu), GGC (α-Gly216) to GGG (Gly), From TTA (α-Leu217) to GTA (Val), AGC (α-Ser219) AAC (Asn), GTG (α-Val224) to CTG (Leu), GTG (α-Val224) to ATG (Met), GTG (α-Val224) to TTG (Leu), GTC (α-Val226) to GCC ( Ala), GCG (α-Ala 231) to ACG (Thr), TTT (α-Phe 233) to CTT (Leu), GGC (α-Gly 236) to AGC (Ser), GAG (α-Glu 240) to GAA (Glu) CAG (α-Gln242) to CAA (Gln), ATG (α-Met257) to GTG (Val), ATG (α-Met257) to ACG (Thr), CTG (α-Leu268) to CTA (Leu), TAT (Α-Tyr271) to TGT (Cys), ACT (α-Asn288) to A C (Asn), GTG (α-Val301) to GTA (Val), ATG (α-Met306) to CTG (Leu), ATG (α-Met306) to TTG (Leu), GCT (α-Ala309) to GCC ( Ala), ATT (α-Ile 314) to GTT (Val), CTG (α-Leu 318) to TTG (Leu), CAG (α-Gln 337) to CAA (Gln), TTC (α-Phe 339) to GTC (Val) , CGC (α-Arg346) to CGG (Arg), ACC (α-Thr350) to GCC (Ala), GCC (α-Ala376) to GCT (Ala), GTT (α-Val423) to ATT (Ile), CGC (Α-Arg425) to CGT (Arg), CCG (α-Pro430) to TCG Ser), AAC (α-Asn447) to AAT (Asn), CCG (α-Pro450) to CCA (Pro), GCG (α-Ala460) to GCA (Ala), GTG (α-Val461) to GGG (Gly) , GAA (α-Glu462) to GAG (Glu), AAC (α-Asn468) to AAT (Asn), ACC (α-Thr470) to GCC (Ala), AGC (α-Ser481) to AGT (Ser), AAT (Α-Asn489) to AGT (Ser), ACC (α-Thr499) to GCC (Ala), TAC (α-Tyr502) to CAC (His), CTC (α-Leu509) to TTC (Phe), CTC (α -Leu509) to TTT (Phe), TTC (α-Phe513) to CTC (L u), AAC (α-Asn520) to AGC (Ser), CGC (α-Arg533) to GGC (Gly), GTT (α-Val549) to GCT (Ala), ACC (α-Thr553) to ACG (Thr) TAA (stop α) to CAA (Gln), TAA (stop α) to GAA (Glu),
2. In β-subunit: CAA (β-Gln2) to CGA (Arg); TTT (β-Phe11) to TTA (Leu), TTT (β-Phe11) to TTC (Phe), CTG (β-Leu13) to CCG (Pro), AAAβ-Lys14) to AGA (Arg), GGG (β-Gly19) to GAG (Glu), GAT (β-Asp24 to GGT (Gly), GAT (β-Asp24) to GAA (Glu), GAA (Β-Glu25) to GAG (Glu), GCC (β-Ala27) to TCC (Ser), GAA (β-Glu29) to GAG (Glu), ACT (β-Thr45) to GCT (Ala), GCG (β -Ala53) to GTG (Val), AAA (β-Lys56) to AGA (Arg), CTG (β-Leu) 8) to CTT (Leu), GAA (β-Glu64) to GAG (Glu), CTT (β-Leu67) to CTC (Leu), CGG (β-Arg70) to CGA (Arg), GCC (β-Ala88) To GCT (Ala), GAT (β-Asp111) to GAA (Glu), CTG (β-Leu113) to CCG (Pro), TCT (β-Ser122) to CCC (Pro), GAG (β-Glu130) to GGG (Gly), CCG (β-Pro152) to ACG (Thr), CCG (β-Pro152) to TCG (Ser), AAC (β-Asn155) to AGC (Ser), AAC (β-Asn155) to AAG (Lys) ), AAA (β-Lys166) to AGA (Arg), AAA (β-Lys173) to G A (Glu), GAC (β-Asp181) to GGC (Gly), CCC (β-Pro184) to CCT (Pro), and 3. In the γ-subunit: AAA (γ-Lys4) to AAG (Lys), ATC (γ-Ile21) to ACC (Thr), AAA (γ-Lys27) to AGA (Arg), GAG (γ-Glu35) to AAG (Lys), ATC (γ-Ile49) to ACC (Thr), ACC ( γ-Thr53) to GCC (Ala), ACC (γ-Thr53) to TCC (Ser), ACC (γ-Thr53) to TGT (Cys), CAT (γ-His67) to TAT (Tyr), AAT (γ- Asn72) to AGT (Ser), CAG (γ-Gln101) to CGG (Arg), ACC (γ-Thr114) TCC (Ser), ACC (γ-Thr114) to GCC (Ala), GCC (γ-Ala122) to GTC (Val), GCG (γ-Ala128) to GTG (Val), and CTG (γ-Leu137) to CTA (Leu).

Similar nucleotide and amino acid mutations can be made in other B _12-dependent dehydratase other than wild-type GDH. By arranging a certain B _12-dependent dehydratase and GDH of interest, location of the exact nucleotide / base requiring mutant are shown.

  The present invention encompasses not only the specific mutations described above, but also those that allow substitution of chemically equivalent amino acids. Therefore, for example, when an amino acid is substituted with alanine, which is an aliphatic nonpolar amino acid, it is expected that the same site may be substituted with serine, which is a chemically equivalent amino acid.

Protein engineering technology of dehydratase It is now possible to try to denature many properties of proteins by combining 3D structural information and classical protein chemistry by means of genetic engineering techniques and molecular graphics methods, ie protein engineering techniques It is. This approach to obtaining an enzyme with altered activity relies first on the generation of a model molecule or the use of a known construction with a sequence similar to the unknown construction.

For this purpose, the three-dimensional crystal structure of the substrate-free form of B _12- dependent glycerol dehydratase was previously determined by X-ray crystal structure analysis (Liao et al., J. Inorganic Biochem. The structure of the enzyme in a complex with 2-propanediol has also been reported (Yamanishi et al., Eur. J. Biochem., 269: 4484-4494 (2002)). These B ₁₂ -dependent three-dimensional models of glycerol dehydratase, and where the “hot spots” of mutations show improved activity in reaction rate (reduced suicide inactivation rate) are typical of these dehydratase enzymes Can be targeted to regions within the dehydratase structure, where alterations in the structure alternative may result in the desired changes in protein properties. Thus, for example, regional site-induced mutagenesis targeting the following wild-type residues is expected to result in additional dehydratase mutants with improved catalytic activity.
1) residues 62-70 (including the second α helix from the N-terminus of the α-subunit) and / or 2) residues 219-236 (region near the active site, the fourth β of the TIM barrel) -Including the chain part followed by a loop of the α-subunit and a short helix).

Recombinant Dehydratase Expression in Host Cells Suitable host cells for recombinant production of 3-HP and 1,3-propanediol by expression of a gene encoding dehydratase may be either prokaryotic or eukaryotic. Limited only by their ability to express active enzymes. Preferred hosts are typically Citrobacter, Enterobacter, Clostridium, Klebsiella, Aerobacter, Lactobacillus, Rospercyl, Aspergillus Schizosaccharomyces, Zygosaccharomyces, Pichia, Kluyveromyces, Candida, Hansenula, Dvariemy D or), Toluropsis, Methylobacter, Escherichia, Salmonella, Bacillus, Streptomyces, and 3-Pseudomonas (P) or Pseudomonas (P) It is useful for producing diols. More preferred in the present invention is Escherichia coli, E. coli. These are E. blattae, Klebsiella, Citrobacter, and Aerobacter.

  Vectors suitable for cloning, transformation and expression of genes encoding suitable dehydratases in host cells, transformation methods, and expression cassettes are well known to those skilled in the art. Suitable vectors are compatible with the bacteria used as host cells. Accordingly, suitable vectors can be derived from, for example, bacteria, viruses (such as phages derived from bacteriophage T7 or M-13), cosmids, yeasts or plants. Protocols for obtaining and using such vectors are known in the art (Sambrook et al., Ibid).

  Vectors or cassettes typically contain sequences that direct the transcription and translation of the relevant gene, selectable markers, and sequences that allow for autonomous replication or chromosomal integration. A suitable vector comprises a 5 'region of a gene containing transcription initiation control and a 3' region of a DNA fragment that controls transcription termination. Most preferably, both control regions are derived from genes homologous to the transformed host cell, but such control regions are not necessarily derived from genes specific to the particular species selected as the production host. It is understood that it is good.

There are a number of initiation control regions (or promoters) useful for driving the expression of the gene of the invention in the desired host cell and are familiar to those skilled in the art. CYC1, HIS3, GAL1, GAL10, ADH1, PGK, PHO5, GAPDH, ADC1, TRP1, URA3, LEU2, ENO, TPI (useful for expression in Saccharomyces), AOX1 (useful for expression in Pichia) And these genes, including but not limited to lac, trp, λP _L , λP _R , T7, tac, and trc (useful for expression in E. coli) Virtually any promoter that can drive is suitable for the present invention.

  Termination control regions may also be derived from various genes that are specific to the preferred host. In some cases, a termination site may not be required, but is most preferably included.

  Once suitable cassettes are constructed, they are used to transform suitable host cells. Introduction of a cassette containing a gene encoding a mutant dehydratase into a host cell can be accomplished by transformation (eg using calcium permeabilized cells, electroporation) or by transfection using a recombinant phage virus. (Sambrook et al., Ibid.).

Industrial production through fermentation The fermentation broth for use in the present invention must contain a suitable carbon substrate. Suitable substrates are well known to those skilled in the fermentation sciences. Preferred carbon substrates are glycerol, dihydroxyacetone, monosaccharides, oligosaccharides, polysaccharides, and monocarbon substrates. More preferred are sugars (eg glucose, fructose, sucrose) and monocarbon substrates (eg methanol and carbon dioxide). Most preferred is glucose.

In addition to a suitable carbon source, the fermentation broth is suitable as known to those skilled in the art, suitable for growth of the culture and promotion of the enzymatic pathways required for 3-HP and 1,3-propanediol production. It must contain minerals, salts, cofactors, buffers, and other components. Particular attention is paid to Co (II) salts and / or vitamin B ₁₂ or their precursors.

  Typically the cells were cultured at 30 ° C. in a suitable culture medium. Preferred growth cultures of the present invention are common commercial media such as Luria Bertani (LB) broth, Sabouraud Dextrose (SD) broth, or yeast malt extract (YM) broth. The culture solution prepared in Other specific or synthetic culture media may be used, and suitable culture media for the growth of specific microorganisms are known to those skilled in the art of microorganisms or fermentation sciences.

  It has been considered that the present invention is implemented using a batch, fed-batch or continuous process, and that any known fermentation mode is suitable (Block et al. ) Furthermore, it was considered that the cells were fixed on a substrate as a whole cell catalyst and exposed to fermentation conditions for producing 3-HP or 1,3-propanediol.

Improved B _12-dependent dehydratase mutants, regardless of the carbon substrate to be used, is expected to be useful for the production of 3-HP and 1,3-propanediol in a variety of processes (U.S. No. 5686276, glycerol to 3-HPA). Recombination of the strain can be swapped out from the dehydratase described in the plasmid of US patent application Ser. No. 10 / 420,587 (DuPont 2002), which is incorporated herein by reference.

  The following examples further illustrate the present invention. It is understood that these examples are provided for illustration only, while illustrating preferred embodiments of the present invention. From the above discussion and these examples, those skilled in the art will be able to ascertain the essential characteristics of the present invention, and various changes and modifications of the present invention may be made without departing from the scope and spirit of the present invention. Can be adapted.

General method:
The procedures required for PCR amplification, DNA modification with endo- and exonucleases (to generate the desired ends for DNA cloning and ligation), and bacterial transformation are well known in the art. . Here, standard molecular cloning techniques are used, and Sambrook J. et al. Fritsch E .; F. , And Maniatis T. et al. "Molecular Cloning: A Laboratory Manual", Second Edition, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York (NY), (1989). (Hereinafter referred to as maniatis); J. et al. Bennan, M .; L. And Enquist L. W. Experiments with Gene Fusions, Cold Spring Harbor Laboratory: Cold Spring, New York (NY), 1984), and Ausubel et al., Molecular Biology (Current Protocols in Molecular Biology) (Greene Publishing and Wiley-Interscience; 1987).

Materials and methods suitable for maintaining and growing bacterial cultures are well known in the art. Techniques suitable for use in the following examples are described below.
1. ) “Manual of Methods for General Bacteriology”, Phillip Gerhardt, R .; G. E. Murray (R.G.E. Murray), Ralph N. Ralph N. Costilou, Eugene W. Negene (Eugene W. Nester), Willis A. Wood (Willis A. Wood), Noel R. Noel R. Krieg, and G. Edited by G. Briggs Phillips, American Society for Microbiology, Washington DC (Washington, D.C.) (1994) or 2. ) Block T. D. Biotechnology: A Textbook of Industrial Microbiology, 2nd Edition, Sinauer Associates: Sunderland, Mass. (1989).
All reagents, restriction enzymes and materials used for bacterial cell growth and maintenance are unless otherwise noted, Aldrich Chemicals (Milwaukee, WI), Milwaukee, Wis., Difco, Detroit, Michigan. Laboratories (DIFCO Laboratories (Detroit, MI)), Gibco / BRL (Gibco / BRL (Gaithersburg, MD)), Gaithersburg, MD, Sigma Chemical Company (StMO, St. Louis, MO) ) Or Promega (Madison, Wis.). PCR reactions were performed on the GeneAMP PCR system 9700 using Amplitaq or Amplitaq gold enzymes from PE Applied Biosystems (Foster City, CA), Foster City, CA unless otherwise noted. . Cycle conditions and reactions were standardized according to the manufacturer's instructions unless otherwise noted herein.

  Unless otherwise noted, the DNA sequencing reaction uses an Expand High Fidelity PCR System from Roche Applied Science (Indianapolis, IN), Indianapolis, IN. Performed on an ABI 377 automated sequencer from PE Applied Biosystems or a PTC-200 DNA engine from MJ Research, Waltham, Massachusetts (MJ Research, Waltham, Mass.). Similarly, data are from the Vector NTI program from Informax, Inc. (Bethesda, MD), Bethesda, Maryland, or the DNA star from DNAStar, Madison, Wisconsin (DNASTAR Inc. (Madison, WI)). Managed using the program.

  The meanings of the abbreviations are as follows. “Sec” means seconds, “min” means minutes, “hr” means hours, “d” means days, “μL” means microliters, “mL” means Means milliliters, “L” means liters, “mm” means millimeters, “μm” means micrometers, “nm” means nanometers, “mM” means millimolar ), “ΜM” means micromolar, “nM” means nanomolar, “M” means molar concentration, “mmol” means millimole, “μmol” means micromole, “ng” means nanogram, “mg” means milligram, “g” means gram, “kB” means kilobase. And, "mU" means millimeter, "U" means a unit.

Strains, vectors and culture conditions Escherichia coli BL21 (DE3) cells were used for enzyme overexpression (Shuster, B. and Rety, J., FEBS Lett.). 349: 252-254 (1994)). Escherichia coli XL1-Blue cells were purchased from Stratagene, La Jolla, Calif. (Stratagene, La Jolla, Calif.). Wild-type Escherichia coli 5K was first found at the University of New Haven, Connecticut, E.coli Genetic Stock Center (CGSC # 4510; Yale University, New Haven, Conn.). ) And lysogenized with lambda DE3 (5K (DE3)). The vector pBluescript II SK + was purchased from Stratagene.

  All kits for molecular biological applications were used according to manufacturer's instructions unless otherwise noted.

Example 1
Construction of an “Xba-library”: random mutagenesis targeting the α-, β-, and γ-subunits of GDH Using error-prone PCR amplification, Klebsiella pneumoniae dhaB1, dhaB2, and dhaB3 A random mutant library targeting the gene was created. Representative sequence analysis of the library demonstrated that there were approximately 4.2 mutations per kB, and enzyme activity measurements determined that approximately 15-25% of the mutants in the library were active. .

Error-prone PCR amplification Emptage et al. (WO 01/12833) describe the construction of plasmid pDT2 comprising the Klebsiella pneumoniae dhaB1, dhaB2, and dhaB3 genes (SEQ ID NO: 1). Plasmid pGD20 is constructed by inserting the HindIII / XbaI fragment of dDT2 (containing dhaB1, dhaB2, and dhaB3) into pBluescript II SK + (Stratagene) that keeps the expression of the GDH gene under the control of the T7 promoter. It was.

A randomly generated mutant library was created that targeted all three GDH genes. First, a sequence comprising dhaB1 (1668 bp), dhaB2 (585 bp) and dhaB3 (426 bp) was amplified from pGD20 by error-prone PCR using the following primers: DHA-F1 (SEQ ID NO: 5) and DHA-R1 (SEQ ID NO: 6). Error-prone PCR was performed using a Clontech mutagenesis kit from Clontech Laboratories, Inc. (Palo Alto, Calif.), Palo Alto, California. The reaction mixture consisted of: 38 μL PCR grade water, 5 μL 10 × AdvanTaq Plus buffer, 2 μL MnSO ₄ (8 mM), 1 μL dGTP (2 mM), 1 μL 50 × Diversified dNTP mix (Diversify dNTP Mix) 1 μL primer mix, 1 μL template DNA, and 1 μL AdvanTaq Plus polymerase. Thermal cycling reactions were performed according to manufacturer's instructions. The 2.7 kB PCR product was digested with Hind III / Xba I and prepared for ligation.

Mutant Library Construction Hind III and Xba I digestion can be used to remove the entire insert containing all three genes of GDH from the pGD20 construct, while the insert size is approximately 2.7 kB, while the vector size Is about 2.9 kB. To facilitate the separation of these two fragments on an agarose gel, pGD20 was digested using Hind III, Xba I and Pst I. Pst I does not cut the vector, but instead cuts the insert at only three places, resulting in four small fragments from the insert. The digested vector therefore moves around 2.9 kB on an agarose gel without contamination from other DNA fragments. The Hind III / Xba I / Pst I digested vector and the Hind III / Xba I-digested error-prone PCR product are then ligated. After ethanol precipitation, the ligation mixture was ready for transformation.

Transformation of the ligation mixture Since the T7 promoter was used for the mutant library, E. coli cells lysogenized with lambda DE3 were used as host cells for mutant enzyme expression. Specifically, 5K (DE3) E. coli strain was used for construction of mutant library. Initially, electroporation-qualified 5K (DE3) cells were made as follows. To 500 mL LB broth in a 2 L sterile flask was added 2.5 mL cells cultured overnight. The culture was incubated at 37 ° C. on a shaker until the OD ₆₀₀ reached 0.5-0.8. Cells were then incubated on ice for 10 minutes followed by centrifugation at 4 ° C. for 10 minutes. After the cells were washed once with 500 mL of ice cold water, the cells were resuspended in 1-2 mL of 10% ice cold glycerol. Aliquots (50 μL) were made in sterile Eppendorf tubes and immediately frozen in dry ice. Competent cells were stored at -80 ° C.

  For transformation, 1 μL of ligation mixture was added to 40 μL of competent cells and the sample was transferred into an electroporation cuvette leaving a 0.1 cm gap. A voltage of 1.7 kv / cm was used for electroporation. Cells were placed on LB plates in the presence of ampicillin and incubated overnight at 37 ° C.

DNA Sequence Analysis of “Xba” Mutant Library Nine mutant colonies were randomly picked for DNA sequencing analysis. After sequencing, the number of mutations that occurred, the mutation location, and the particular type of mutation observed were analyzed for each mutant. Analysis revealed that all types of base substitutions were present in the mutant strain, suggesting a lack of bias for specific mutation types. In addition, only one deletion mutation was observed in the 9 mutant clones analyzed, and no base insertion mutation was identified. This suggested that the frequency of deletions and insertions in the mutant library was very low. The average mutation rate was 4.2 points mutation per kB. GDH activity measurements showed that about 15-25% of the mutants in the library were active.

Example 2
Construction of “Sma-library”: Random mutagenesis targeting the α-subunit of GDH and part of the β-subunit. A random mutant library was created. Representative sequence analysis of the library demonstrated that there are approximately 4.5 mutations per kB. Enzyme activity measurement determined that about 15-25% of the mutants in the library were active.

Error-prone PCR amplification and mutant library construction Using the following primers, the entire dhaB1 gene and an approximately 200 bp portion of the dhaB2 gene were amplified by error-prone PCR using pGD20 as a template. DHA-F1 (SEQ ID NO: 5) and DHA-R2 (SEQ ID NO: 7). Error-prone PCR reactions were performed using the Clontech mutagenesis kit as described in Example 1. The 1.9 kB PCR product was then digested with Hind III and Sma I.

  To prepare the vector, the pGD20 plasmid was digested with Hind III and Sma I (removing the wild type dhaB1 gene and part of the dhaB2 gene) and then purified from an agarose gel. The Hind III / Sma I-digested error-prone PCR product and the Hind III / Sma I-digested pGD20 vector were ligated. The ligation mixture was transformed into E. coli 5K (DE3) by electroporation in the same manner as described in Example 1 for creating a mutant library.

DNA Sequence Analysis of “Sma” Mutant Library Ten mutant colonies were randomly picked for DNA sequencing analysis and examined for library integrity. The number of mutations that produced each mutant, the mutation location, and the specific type of mutation observed were determined. Based on these results, the average mutation rate in the Sma-library was determined to be 4.5 point mutations per kB. There was no obvious bias for any particular mutation type, and the frequency of deletions and insertions in the mutant library was very low. Enzyme measurements revealed that approximately 15-25% of the mutants in the library were active.

Example 3
Regional targeting amino acids 141-152 (“PpuMI-library”), 219-226 (“4BR1-library”) and 330-342 (“RsrII-library”) of the α-subunit of GDH Random mutagenesis Crystal structure of GDH (Liao et al., J. Inorganic Biochem., 93 (1-2): 84-91 (2003); Yamanishi et al., Eur. J. Biochem. 269: 4484 -4494 (2002)), the following regions of the α-subunit of GDH were targeted for regional random mutagenesis. 1) amino acid numbers 141-152, 2) amino acid numbers 219-226, and 3) amino acid numbers 330-342. Since each of these regions is rather short in length, an oligo-induced mutagenesis approach was used to create these three mutant libraries. This involves a multi-step process that first creates silent mutations corresponding to unique restriction sites upstream or downstream of each region to be mutated to facilitate cloning. Denatured oligonucleotide primers were prepared in this way and used in PCR reactions to mutagenize at the target region of the α-subunit. These mutagenized PCR fragments were then cloned into E. coli to create a “PpuMI-library”, “4BR1-library”, and “RsrII-library”.

Introducing Silent Mutations into pGD20 To create a unique restriction site in plasmid pGD20, one silent mutation was generated upstream or downstream of each target region. The following primer pairs were used to create point mutations in each region. Nucleotides shown in bold capital letters in each primer indicate the position at which a particular mutation is introduced.

  Perform mutagenesis experiments using Stratagene quick-change (QuikChange) site-induced mutagenesis kits from Stratagene (La Jolla, CA), La Jolla, California, according to manufacturer's instructions did. Following mutagenesis, plasmids were purified from each mutant clone. Point mutations were confirmed by restriction enzyme digestion followed by direct DNA sequence analysis.

Oligo-induced mutagenesis Three modified oligonucleotides were synthesized (pGD20RM-F3, TB4B-R1, and pGD20RM-R4 shown below) to produce a regional random mutant library. Standard conditions were used for oligonucleotide synthesis for nucleotides shown in capital letters. In contrast, nucleotides shown in bold lowercase used the degenerate nucleotide mixture shown below each primer during synthesis. Thus, it was predicted that a 1-2 point mutation would result in each “denaturing region” of the primer.

pGD20RM-F3 (for the aa 141 to aa 152 region— “PpuMI library”):
5′-GCC CGC AGG ACC CCC TCC aac cag tgc cac gtc acc aat ctc aaa gat aat ccg GTG CAG ATT-3 ′ (SEQ ID NO: 15)
94% A mixed with a = 2% G, 2% C, and 2% T,
94% G mixed with g = 2% A, 2% C, and 2% T,
94% C mixed with c = 2% G, 2% A, and 2% T,
t = 2 94% T mixed with 2% G, 2% C, and 2% A.

TB4B-R1 (for region aa 219 to aa 226— “4BR1 library”):
5′-GCG TGG GTT AAC cag cta cgc cga gac ggt gtc ggt cta cGG CAC CGA AGC GGT ATT TAC C-3 ′ (SEQ ID NO: 16)
94% A mixed with a = 2% G, 2% C, and 2% T,
94% G mixed with g = 2% A, 2% C, and 2% T,
c = 2 4% A, 2% T, and 94% C mixed with 2% G,
t = 2 94% T mixed with 2% A, 2% G, and 2% C.

pGD20RM-R4 (for aa 330 to aa 342 region-"RsrII library"):
5′-GGT GCG CGC GGT GCG GCG AAT ATC cga gtg gga gaga agt ctg gtt gtt ggg gga cgc cac tttc GAG GTC GAG-3 ′
95% A mixed with a = 1.66% G, 1.66% C, and 1.66% T;
95% G mixed with g = 1.66% A, 1.66% C, and 1.66% T;
c = 1.66% G, 1.66% A, and 95% C mixed with 1.66% T;
t = 1 95% T mixed with 1.66% G, 1.66% C, and 1.66% A.

A high fidelity PCR reaction is then performed and mutagenic PCR fragments are generated using the following primer pairs:
PpuMI-library: pGD20RM-F3 and DHA-R2 (SEQ ID NOs: 15 and 7),
4BR-1 library: TB4B-R1 and GD-C (SEQ ID NOs: 16 and 14),
RsrII-library: DHA-F1 and pGD20RM-R4 (SEQ ID NO: 5 and 17).

  PCR fragments were purified from agarose gels and then PpuM I / Xba I (for PpuMI-library), Hpa I / Xba I (for 4BR-1 library), and Hind III / Rsr II (RsrII-live) Digested for each rally).

Mutant library construction. Mutant pGD20 construct (in which a unique restriction enzyme digestion site of PpuM I, Hpa I or Rsr II is introduced by site-induced mutagenesis) Digested with Hind III / Rsr II, respectively. The linearized vector was purified from an agarose gel and then ligated to the restriction enzyme-digested PCR product. The ligation mixture was electroporated into E. coli strain 5K (DE3) as described in Example 1 to prepare a mutant library.

Sequencing analysis for regional libraries 9-10 mutant colonies were randomly picked for DNA sequencing analysis from each regional library. In the PpuMI-library, the average mutation rate per mutant was 2.8 mutations (n = 9). In the 4BR1-library (n = 8), the average mutation rate per mutant was 2.0 mutations. Finally, sequencing data from the Rsr II library (n = 10) revealed an average mutation rate of 2.2 mutations per mutant. No insertions or deletions were observed in any of the three libraries. All types of base substitutions were detected, suggesting a lack of bias for any particular mutation type.

Example 4
Screening assay development A screening assay was developed that allows the B _12- dependent dehydratase reaction to be “initiated” manually by the addition of B ₁₂ coenzyme and substrate (glycerol). The dehydratase reaction product 3-HP is detected by a colorimetric aldehyde assay. Using this assay, the preliminary analysis related to aging of a typical B _12-dependent dehydratase reaction in the presence of glycerol and 1,3-propanediol, the initial reaction velocity v, and were observed enzyme not The rate of activation k allows the development of a rapid theoretical technique to estimate the _{observed inactivity} .

If no source of coenzyme B ₁₂ is a cofactor of the screening assays principle enzyme, it is cultured to express either wild-type B _12-dependent dehydratase or mutant B _12-dependent dehydratase, apo -B _12- dependent dehydratase is produced. In contrast, holoenzymes spontaneously form when coenzymes are added to toluene- and detergent-permeabilized whole cells. Therefore, it is possible to “initiate” the B _12- dependent dehydratase reaction by adding coenzymes and substrates to such cells. The reaction product 3-HP can be detected by a colorimetric aldehyde assay using the reagent 3-methyl-2-benzothiazolinone (MBTH) and the oxidizing agent ferric chloride (Zurek). ) G., and Karst U., Analytica Chimica Acta, 351: 247-257 (1997)).

Preliminary assay to investigate time course of typical GDH response (with or without 1,3-propanediol) Cells producing wild type GDH were treated with 15% Lennox Bros (Gibco-BRL, Gaithersburg, MD). (Prepared by diluting 15 volumes of Lennox broth from Gibco-BRL (Rockville, MD) with 85 volumes of 0.5% NaCl followed by sterilization) in New Brunswick, NJ Shake (250 rpm) in an Innova 4300 incubator from Brunswick Scientific (New Brunswick Scientific, NJ) and grow overnight at 37 ° C. Cells were permeabilized as follows. Aliquots of culture (0.3 mL) were dispensed into square wells of polypropylene deep well plates from Beckman-Coulter (Fullerton, Calif.), Fullerton, Calif., 2.5% (v / v) 10 μL of toluene containing a Triton X-100 detergent was placed in each well. The plates were then shaken at maximum speed for 10 minutes on an IKA MTS4 shaker from IKA-Werke GmbH (Staufen, Germany), Staufen, Germany. Working under red light to protect the coenzyme B _12, the reaction was initiated by addition of substrate solution 5 volume permeabilized cell suspension 1 volume. The substrate solution contained ₁₂ mM glycerol and 24 μM coenzyme B ₁₂ in 0.1 M K-Hepes (pH 8). At regular intervals, a 12.5 μl aliquot of the reaction was placed in a well of a 96-well plate containing 3 mg / mL MBTH in 12.5 μl 0.4 M glycine-HCl (pH 2.7). At least 20 minutes after sampling, 125 μL of 5.5 mM FeCl ₃ in 10 mM HCl was added to each of the MBTH-containing samples. After an additional 20 minutes or more, the blue color associated with GDH activity is measured at 670 nm using a Spectramax 160 plate reader from Molecular Devices (Sunnyvale, Calif.), Sunnyvale, Calif. As quantified. A similar experiment was performed with the substrate solution supplemented with 50 mM 1,3-propanediol.

Theoretical analysis of assay results The data from the time course response was plotted as time-versus-OD ₆₇₀ as shown in FIG. Specifically, shown Figure 1 is above the curve, GDH with 10mM glycerol was used as a substrate _{(K m} about 0.5 mM) (room temperature, in 0.1M K-HEPES, pH8) the time course of reaction. As inactivation occurs over time, the rate of product formation decreases rapidly. The theoretical curve passing through the points is drawn by applying the three parameters to the following equation:
y = T ₀ + amp (1-exp (−k _{observed inactivity} ^* time))
These parameters are
1. ) Background (assay t = 0 (T ₀ ) value),
2. 2.) Amplitude value (amp) of the limited total absorbance that occurs during the assay, and ) First-order inactivation rate constant (k _observed inactivity) that controls curvature.
The fitting parameters are related to the rate characteristics of the reaction as follows. The amplitude is equal to the initial reaction rate (v) divided by the _observed enzyme inactivation rate (k _{observed inactivity} ). The initial speed is [GDH] [gly] k _cat / (K _m + [gly]). Subtracting the assay background, further analysis reveals:
1. Both the v and k _{observed inactivity} can be estimated from only two samples taken during the reaction, the initial point for estimation of v (T1) and the end point for amplitude (T2),
2.1 / k The _{observed inactivity} can be estimated as T2 / T1.

The lower curve in FIG. 1 shows the effect of including 50 mM 1,3-propanediol (K _i ˜15 mM) in the assay. The initial reaction rate decreased by about 20%, but inactivation was about 3 hours faster. This is because the competitive inhibition is relatively small, but the inactivation rate constant of the enzyme-1,3-propanediol complex (k _{-inactive 1,3-propanediol} ) is that of the enzyme-glycerol complex ( k _{inactive glycerol} ).

  A rate scheme incorporating both deactivation processes is shown below.

Example 5
Two-point high throughput screening assay Based on the methodology described in Example 4, the mutant colonies prepared in Examples 1, 2, and 3 were examined using a high throughput screening assay protocol. This assay specifically measured GDH reaction products at two time points during the assay. T1 was measured in 30 seconds after _{T 0,} T2 were measured after 40 minutes of _{T 0.}

  Colonies from the plate-cultured library were picked into 94 wells of a standard 96-well microtiter plate containing 15% Lennox broth at 0.15 mL / well. The remaining two wells were inoculated with cells producing wild type GDH and the cells were cultured in a static incubator for 4-6 hours at 37 ° C. As an alternative, if it was desired to store the picked up cells at −80 ° C. prior to screening, the medium containing glycerol (10% v / v) and the picked up cells were cultured overnight prior to storage. Prior to screening, pre-frozen cells in 96-pin inoculators from V & P Scientific (San Diego, Calif.), San Diego, Calif. Were used to obtain fresh 96 containing 15% Lennox broth without glycerol. -Transfer to a well plate and incubate without shaking at 37 ° C for 6-16 hours.

  For GDH mutant screening, the cell culture protocol was as follows. 15% Lennox broth medium was dispensed (0.3 mL / well) into the wells of a polypropylene deep well 96-plate from Beckman-Coulter (Fullerton, Calif.), Fullerton, California. Using a 96 pin-long pin inoculator (V & P Scientific), cells were inoculated from shallow 96-well plates into deep wells to provide 3M Healthcare (3M Health, St. Paul, Minn.). The plate was covered with 3 inch wide micropore surgical tape from Care (St. Paul, MN). Shallow 96-well plates were stored at 4 ° C. until the assay was completed, and they were used as a source of live cells to further investigate lines identified as potentially improved. Cells in deep well plates were cultured overnight at 37 ° C. with shaking (250 rpm). The air in the Innova 4300 incubator from New Brunswick Scientific (New Brunswick, NJ) was humidified by a wet sponge in a plastic tray.

Cells were permeabilized in 96-well plates by adding 10 μL of toluene containing 2.5% (v / v) Triton X-100 detergent to each well. The plates were then shaken at maximum speed for 10 minutes on an IKA MTS4 shaker from IKA-Werke GmbH (Staufen, Germany), Staufen, Germany. Bio in a Plexiglas® enclosure from Rohm & Haas (Philadelphia, PA) covered with a red plastic film to protect the substrate mixture from white room light Using a Biomek 2000 robot (Beckman-Coulter), aliquots (8 μL) of permeabilized cells were transferred to 96-well reaction plates in 0.1 M potassium-HEPES buffer. 24μM coenzyme _B 12 in (pH 8), 12 mM glycerol, and substrate mixture of 40μL containing 50 mM 1,3-propanediol (be prepared under red light, in a container wrapped foil until needed - (Stored at 0 ° C.) was added to the cells by robot to initiate the reaction at room temperature 30 seconds after substrate addition, an aliquot of the 12.5 μL reaction (T1 sample) was 0.4 M glycine-HCl (pH 2. 7) 12.5 μL of 3-methyl-2-benzothiazolinone (MBTH) in 7) was transferred to the second plate containing the wells, approximately 40 minutes after addition of the substrate, to the third plate containing MBTH. An aliquot of the same reaction (T2 sample) was transferred, and at least 20 minutes after this transfer, a 5.5 mM FeCl ₃ solution in 125 μL of 10 mM HCl was added to each of the second and third plates (containing MBTH). After an additional 20-60 minutes, Molecular Devices (Sunny), Sunnyvale, California. Vale, CA)) was used to quantify the blue color associated with GDH activity as absorbance at 670 nm using a Spectramax 160 plate reader.

  Data from the plate reader was transferred to a modified Excel computer program from Microsoft® (Redmond, WA), Redmond, WA. The program matches the first (T1) and second (T2) samples from each reaction plate, outputs the plate, the position within the plate, and the T1, T2, and T2 / T1 ratios for each reaction. It is organized to be a table of results. Data were examined for samples that showed very high values in the T1, T2, or T2 / T1 ratio. This was facilitated by the program's ability to sort the data by all these parameters. In order to examine the most promising candidates in more detail, streaks were cultured from reserved shallow 96-well plates.

Example 6
Screening GDH Mutant Library and Identifying Positive Hits Using the automated high-throughput assay described in Example 5, nearly 100,000 mutant colonies from 5 mutant libraries were screened. . These libraries included:
1. ) An Xba library targeted to α-, β-, and γ-subunits (DhaB1, DhaB2, and DhaB3) as described in Example 1;
2. ) Sma library targeting α- and a small part of β-subunit as described in Example 2;
3. ) As described in Example 3, a. a. 141-a. a. PpuMI library targeting 152,
4). ) As described in Example 3, a. a. 219-a. a. 4. 4BR1 library targeting 226, and 5. ) As described in Example 3, a. a. 330-a. a. RsrII library targeting 342.

  Every individual isolate was derived from one of the five libraries described above. Most individual isolates were uniquely identified by a label indicating the library source followed by a number (eg, “Xba3010”). All putative “hits” from the first screen were confirmed in the follow-up assay. Most mutagenic effects were classified according to four broad categories based on rate parameters as described below. Sequence analysis of the mutant gene and subsequent comparison with the wild type gene allowed identification of specific point mutations present in each mutant gene.

Mutant Library Screening and Hit Confirmation Following a primary screen of nearly 100,000 mutants, putative hits were confirmed by a follow-up confirmation assay. Briefly, each putative hit was re-assayed in 8 wells. The results from each individual clone were statistically analyzed to obtain the mean and standard deviation of T1 (the amount of aldehyde measured at 30 seconds) and T2 (the amount of aldehyde measured at 40 minutes). These results were compared with the wild type enzyme.

FIG. 2 shows a typical follow-up assay result plotting the number of assays on the x-axis against OD _{670 nm} on the y-axis. Results from each individual assay (n = 8) at T1 and T2 for Xba3010 and wild type clones are presented. The mean value is shown as a horizontal line and is presented numerically in parentheses ± standard deviation. In general, the standard deviation was about 10-15%.

Screening results Table 6 summarizes the follow-up assay results for hits having either a T2 / T1 ratio above wild type, or a T2 above wild type, or both. The T2 / T1 ratio provides an indication of enzyme stability in a 40 minute reaction that occurs in the presence of 1,3-propanediol and glycerol. T2 suggests the total turnover number of the enzyme before it is completely inactivated. The higher the T2 / T1 ratio, the better the enzyme stability. Thus, mutant enzymes with improved stability have a reduced inactivation rate in the presence of glycerol and 1,3-propanediol compared to the wild type enzyme.

In Table 6, the average values of T1 and T2 from the follow-up assay following normalization to wild type results are reported. Most mutant enzymes are classified as type-1, -2, -3, or -4 mutants based on the following definitions.
● Type 1 mutant: The T2 / T1 ratio is improved compared to the wild type, while the absolute value of the T2 value is decreased.
● Type-2 mutant: While both the T2 / T1 ratio and T2 are improved over the wild type, the degree of T2 improvement is less than the improvement of the T2 / T1 ratio.
● Type-3 mutant: both T2 / T1 ratio and T2 are improved over the wild type, and the degree of improvement of both is the same.
● Type-4 mutant: T2 / T1 ratio is equivalent to or lower than that of wild type, but T2 is improved.

  Table 6 also summarizes the specific point mutations identified in each mutant strain. The SEQ ID number of the DNA sequence of the enzyme is listed in the first column of the table.

  As seen in the sequencing results, most mutations are identified as amino acid substitutions (except for silent mutations). However, two of the mutant strains (Sma3002 [SEQ ID NO: 140] and Xba3009 [SEQ ID NO: 179]) are fusion proteins. Specifically, the stop codon (TAA) of the gene encoding the α-subunit (dhaB1) changes to CAA (Gln) or GAA (Glu) in Sma3002 and Xba3009, respectively. Since there is 15 bp between this stop codon and the first codon of the gene encoding the β-subunit (dhaB2), neither of these fusion proteins caused a frameshift in the β-subunit. This allowed both mutant enzymes to maintain activity. The first codon (GTG) of the β-subunit is usually recognized by fMet-tRNA. However, in fusion mutants, this codon must be recognized by Val-tRNA. The Sma3002 fusion protein contained a linker consisting of 6 amino acid residues Gln-Gly-Gly-Ile-Pro-Val (SEQ ID NO: 18). In the Xba3009 fusion protein, the linker was Glu-Gly-Gly-Ile-Pro-Val (SEQ ID NO: 19).

Example 7
Biochemical analysis of mutants using purified enzymes Following the overexpression and purification of each enzyme, a more detailed characterization of 5 mutants and wild type GDH was performed.

Overexpression and purification For further biochemical analysis with the wild type enzyme, Xba3007 (type-1, SEQ ID NO: 40), Sma3002 (type-2, SEQ ID NO: 140), Xba3010 (type-3, SEQ ID NO: 175), Xba3009 ( Type-4, SEQ ID NO: 179) and 4BR1001 (Type-2, SEQ ID NO: 171)) were selected. First, the plasmid was purified from a 5K (DE3) E. coli host strain and transformed into E. coli BL21 (DE3). Prior to the addition of 1.0 mM IPTG, cells were cultured in ampicillin-containing LB medium to an OD ₆₀₀ of 0.6-1.0. After induction at 37 ° C. for 3 hours, the cells were collected by centrifugation and washed once with 20 mM HEPES-KOH buffer (pH 8.0). Cell pellets were stored at -80 ° C.

For enzyme purification, the cell pellet was first prepared in 20 mM HEPES-KOH buffer containing a complete miniprotease inhibitor cocktail tablet from Roche, Palo Alto, Calif. And 0.5 mM EDTA. Resuspended in (pH 8.0). Cells were disrupted by sonication (Branson model 450; 20% power, 50% pulse, 4 minutes in ice bath) followed by centrifugation (40,000 × g, 30 minutes, 4 ° C.). The clear supernatant was spun down (110,000 × g, 4 ° C., 1 hour) and (NH ₄ ) ₂ SO ₄ was slowly added into the supernatant on ice to bring the solution to 50% saturation. The solution was stirred on ice for 25 minutes, followed by centrifugation (40,000 × g, 30 minutes, 4 ° C.). The pellet was resuspended in 2 mL of running buffer (100 mM HEPES-KOH pH 8.2, 100 mM 1,2-propanediol and 1 mM DTT) and then equilibrated with running buffer. Placed in a 16/60 high load Superdex 200 particle size exclusion column from Pharmacia Biotech (Piscataway, NJ), Piscataway.

  The enzyme was eluted with a buffer at a flow rate of 0.25 mL / min, and the eluate was collected using a fraction collector (3 mL / fraction). Fractions were assayed for enzyme activity using the assay described in Example 5, and the active fractions were then pooled and sent to a Centricon YM100 from Millipore, Bedford, Mass. Concentrated using The concentrated enzyme was again passed through the same column using fresh running buffer consisting of 100 mM HEPES-KOH (pH 8.2) and 1 mM DTT. The purified enzyme was 75-95% pure as assessed by SDS-PAGE electrophoresis using a 10-20% gradient gel and Coomassie blue staining.

Biochemical characterization of mutant GDH enzyme:
Detailed enzyme rate analysis using purified wild type and mutant GDH enzymes was performed. The MBTH-colorimetric aldehyde assay (Zurek G., and Karst U., Analytica Chimica Acta, 351: 247-257 (1997)) was used to measure product formation and to determine enzyme activity. seeking the _maximum K _M and V line Weber - it was determined from Burke plot. k _cat was calculated from V _max . The _{K M} and _{k cat} of the enzyme found in Table 7.

Finally, a detailed analysis of the inactivation characteristics of each mutant enzyme was performed. Glycerol inactivation rate constants were measured as described in Example 4. For the air inactivation rate constant, the enzyme was diluted to 27 μg / mL with 0.1 M K-HEPES buffer (pH 8) containing 21 μM coenzyme B ₁₂ and then incubated in air at room temperature. Total turnover was measured after 0.25, 0.5, 1, 2, 5, 10, 15, 20 and 30 minutes of incubation. To measure total turnover, add 10 μL enzyme solution to 190 μL reaction solution containing 200 mM glycerol, 24 mM coenzyme B ₁₂ , and 0.1 M K-HEPES buffer (pH 8). The reaction mixture was incubated at room temperature for 2 hours. Total turnover numbers were estimated by measuring 3-HP concentrations using the MBTH-colorimetric aldehyde assay (Zurek G., and U. (Karst U., supra). Data is plotted as time vs. total turnover number, A = A ₀ exp (−k _air t) (where “A” is the total turnover number, “A ₀ ” is the total turnover number at 0:00) , “K _air ” is the air deactivation rate constant, and “t” is time) to estimate the total turnover number by curve fitting.

To measure 1,3-propanediol inactivation, 0.1 M K containing 21 μM coenzyme B ₁₂ and various concentrations of 1,3-propanediol (ie 1, 20, 100 and 300 mM) -The enzyme was diluted to 27 μg / mL in HEPES buffer (pH 8). The deactivation rate constant was estimated for each 1,3-propanediol concentration by the method used to measure the air deactivation rate constant. The inactivation rate constant was plotted against the 1,3-propanediol concentration, and the maximum inactivation rate constant due to 1,3-propanediol was estimated from the curve. The dissociation constant of 1,3-propanediol was 1,3-propanediol concentration at which the deactivation rate constant was half of the maximum deactivation rate constant.

The analysis results of this inactivation characteristic are shown in Table 8 below. Glycerol or 1,3-propanediol or light either in the absence, but in the presence air (oxygen), B _12-dependent dehydratase holoenzyme enzyme inactivation occurs in a biphasic rate. k _gly is the inactivation rate constant measured in the presence of 10 mM glycerol without 1,3-propanediol. Since all mutants and wild type dehydratase showed two-phase air inactivation, k _{air, F} is the air inactivation rate constant for the rapid phase, and k _{air, S} is the slow phase air inactivation. It is an activation rate constant, and k1,3 _-propanediol is the maximum deactivation rate constant by 1,3-propanediol. K _d is the dissociation constant of 1,3-propanediol.

Example 8
Second mutagenesis with selected first generation mutant combinations In order to further improve the mutants obtained from the first mutagenesis, a second mutagenesis was performed and several first generation mutations were performed. Mutations from strains were combined.

  In order to produce second generation mutants, plasmids from several first generation mutants (eg, Sma3002 or Xba3009) containing multiple mutations were first purified from host cells. These plasmids were then introduced with single point mutations found in Xba3007, Xba3029 or 4BR1001 to generate second generation mutants 2-F4, 12-B1, 13-B7, and 16-H5. Table 9 summarizes details about each of these variants and the primers used to generate them. Nucleotides shown in bold capital letters in each primer indicate the position at which a particular mutation is introduced.

  Mutations as described in Example 3 using the Stratagene Quick-Change Site-Induced Mutagenesis Kit from Stratagene, La Jolla, California (Stratagene, La Jolla, Calif.) Provocation experiments were performed.

  As described in Example 5, the T2 / T1 ratio and T2 value of each second generation mutant were determined. These results are shown in Table 10 below. The SEQ ID number of the DNA sequence of the enzyme is listed in the first column of the table.

Example 9
Construction and analysis of pure fusion mutants Two fusion mutants, Xba3009 (SEQ ID NO: 179) and Sma3002 (SEQ ID NO: 140) were described in Example 6. Both contained other mutations in addition to the fusion itself. In order to investigate the effects of α- and β-fusion, two pure fusion mutants (1E1 and 20G7) were constructed.

Construction of pure fusion mutant The stop codon of α-subunit (TAA) was changed to CAA (1E1) or GAA (22-G7). This modification was achieved by introducing a single point mutation into the wild type GDH plasmid. In order to create a point mutation, the following two pairs of primers were used.
For the 1-E1 mutant:
1-E1-F1: 5′-gac acc att gaa Caa ggc ggt att cct-3 ′ (SEQ ID NO: 26)
1-E1-R1: 5′-agg aat acc gcc ttG ttc aat ggt gtc-3 ′ (SEQ ID NO: 27)
For the 22-G7 mutant:
22-G7-F1:
5′-ccc gac acc att gaa Gaa ggc ggt att cct gtg-3 ′ (SEQ ID NO: 28)
22-G7-R1:
5′-cac agg aat acc gcc ttC ttc aat ggt gtc ggg-3 ′ (SEQ ID NO: 29)

  Nucleotides shown in bold capital letters in each primer indicate the position at which a particular mutation is introduced.

  Mutations as described in Example 3 using the Stratagene Quick-Change Site-Induced Mutagenesis Kit from Stratagene, La Jolla, California (Stratagene, La Jolla, Calif.) Provocation experiments were performed. The T2 / T1 ratio and T2 value of these second generation mutants were determined as described in Example 5. These results are shown in Table 11 below. The SEQ ID number of the DNA sequence of the enzyme is listed in the first column of the table.

Example 10
Analysis of mutations found in mutant Sma3002 Sma3002 (SEQ ID NO: 140), the first generation mutant identified in Example 6, showed a significant improvement in both T2 / T1 ratio and T2 value. It contained 4 point mutations, one of which contained the fusion mutation explored in detail in Example 9. In order to investigate the effects of these mutations individually, two additional mutants (α-Y271C and β-Q2R) were constructed.

  Using the methodology followed, single point mutations were introduced into the wild-type plasmid using uniquely designed primers as shown below. As in the previous examples, the nucleotides in bold capital letters in each primer indicate the position at which a particular mutation is introduced.

  Mutations as described in Example 3 using the Stratagene Quick-Change Site-Induced Mutagenesis Kit from Stratagene, La Jolla, California (Stratagene, La Jolla, Calif.) Provocation experiments were performed. T2 / T1 ratios and T2 values of these mutants were measured as described in Example 5. Table 13 shows the results. The SEQ ID number of the DNA sequence of the enzyme is listed in the first column of the table.

The α-Y271C mutation decreased the T2 value but failed to change the T2 / T1 ratio. This mutation therefore reduces the enzyme's k _cat . Another mutant β-Q2R does not appear to significantly affect either the T2 or T2 / T1 ratio.

  To determine if any of the three non-fusion mutations in Sma3002 (SEQ ID NO: 140) work together to increase the stability of the enzyme in the presence of 1,3-propanediol and glycerol, Three additional mutants were created. In each of these mutants, one of the three non-fusion mutations (α-Y271C, α-Y502H, or β-Q2R) in Sma3002 was removed by introducing the wild type DNA sequence as a single point mutation. From this, three Sma3002-derived mutants each containing two of the original non-fusion mutations present in Sma3002 were obtained. Single point mutations were introduced into the Sma3002 plasmid as described in Example 3 using the Stratagene QuikChange site-induced mutagenesis kit and the primers are shown in Table 14 below. .

  T2 / T1 ratios and T2 values of these mutants were measured as described in Example 5. Table 15 shows the results of this analysis. The SEQ ID number of the DNA sequence of the enzyme is listed in the first column of the table.

Example 11
Third Mutagenesis by Addition of Several First Generation Mutations to Second Generation Mutants To improve the total turnover of GDH enzyme, measured at the value reported here as T2, in the preceding examples Although significant improvements were made, a third mutagenesis was performed. In these reactions, point mutations selected from the first generation mutants were introduced into the two second generation mutants (ie 1-E1 and 8-C9) showing the greatest improvement in T2 values. This was first achieved by purifying the 1-E1 and 8-C9 mutant plasmids from the host cells. Next, using the primers shown in Table 16 below and the Stratagene quick-change (QuikChange) site-induced mutagenesis kit from Stratagene, La Jolla, Calif. (Stratagene, La Jolla, Calif.). As described in Example 3, single amino acid substitution mutations found in Xba3007, Xba3029, or 4BR1001 were introduced into these plasmids.

  The T2 / T1 ratio and T2 value of these third generation mutants were measured as described in Example 5. Table 17 shows the results. The SEQ ID number of the DNA sequence of the enzyme is listed in the first column of the table.

  Overall, the third generation mutants had a significant improvement in enzyme stability. All mutants had improved stability (T2 / T1 ratio) and total turnover (T2) compared to the best first generation mutant (ie, Sma3002 (SEQ ID NO: 140)). In fact, mutants 20-B9 and 21-D10 had T2 values that exceeded all mutants created so far.

Example 12
Biochemical characterization of several second and third generation mutants To further characterize selected second and third generation mutants, using the method described in Example 4, 10 mM Inactivation rate constants of these mutants were determined in the presence of glycerol and 50 mM 1,3-propanediol. These results are summarized in Table 18 below.

  As expected, the results were consistent with the measurement of the T2 / T1 ratio.

  Since one use of improved GDH is 1,3-propanediol bioproduction, it is important to determine the total turnover number of mutants in the presence of high concentrations of 1,3-propanediol. The above assay was performed in the presence of only 50 mM. Later in the fermentation for 1,3-propanediol bioproduction, the 1,3-propanediol concentration can reach 1M. A second two-point assay was therefore performed. Specifically, T2 values were also measured for several promising mutants in the presence of 10 mM glycerol and 600 mM 1,3-propanediol. Table 19 summarizes the results.

The results were somewhat similar to those obtained for T2 _{(50 mM)} , but several mutants performed better than expected. The T2 _{(600 mM) of} Sma3002 was 3.3 times higher than that of the wild type enzyme. Each third generation mutant (15-E4, 18-D7, 20-B9, and 21-D10) showed a further improvement in T2 _{(600 mM)} compared to Sma3002. The results suggested that these third generation mutants with higher T2 _{(600 mM)} were more resistant in higher concentrations of 1,3-propanediol. These mutants are expected to be very useful for 1,3-propanediol bioproduction.

Example 13
One-point high-throughput screening assay for measuring total enzyme turnover in the presence of high concentrations of 1,3-propanediol. If the 1,3-propanediol concentration is higher than about 300 mM, almost immediately after the start of the reaction. Inactivation of GDH occurs. Under these conditions, T1 cannot be accurately measured, so mutants cannot be screened using the high-throughput screening assays described in Examples 4 and 5. Total enzyme turnover in the presence of high concentrations of 1,3-propanediol is one of the important enzyme kinetic parameters that should be improved here. This total enzyme turnover number can be improved by either reducing the inactivation rate or increasing k _cat . A modified high-throughput screening assay was developed to screen for mutants with improved total enzyme turnover in the presence of high concentrations of 1,3-propanediol.

Briefly, the mutant cells were cultured in a 96-well plate and permeabilized as described in Example 5. Using a Biomek 2000 robot from Beckman-Coulter (Fullerton, Calif.), Fullerton, Calif., Aliquots (8 μL) of permeabilized cells were transferred to 96-well reaction plates. did. 24 μM coenzyme B ₁₂ , ₁₂ mM glycerol in 0.1 M potassium-HEPES buffer (pH 8) using Qfill2 from Genetics in New Milton, Hampshire, UK (Geneticx (New Milton, Hampshire, UK)), and 40 μL of substrate containing 720 mM 1,3-propanediol was added to the cells to initiate the reaction at room temperature. Plates were incubated at room temperature for approximately 70 minutes before using a Biomek 2000 robot (Beckman-Coulter) in 12.5 μL of 0.4 M glycine-HCl (pH 2.7). A 12.5 μL aliquot of the reaction was transferred to a second plate containing 3-methyl-2-benzothiazolinone (MBTH), which gives an accurate measure of total turnover with a reaction time of 70 minutes. The concentration of the product 3-HP was determined as described in Example 5. Spectramax from Molecular Devices (Sunnyvale, Calif.), Sunnyvale, Calif. ) 67 using a 160 plate reader The absorbance at nm was measured to estimate the total enzyme turnover number (T (600)).

  Since Qfill2 can add substrate to a single 96-well plate within 15 seconds, this single point colorimetric assay is easy to perform. Furthermore, the assay can provide a higher throughput capacity compared to the assays described in Examples 4 and 5. Table 20 shows the results of a comparison of the assays performed in Example 5 and this example for wild-type GDH and several mutants with different enzyme rate parameters.

These results demonstrate that each assay is screened for different rate parameters. Variants identified by the screening assay described in this example are more resistant to higher concentrations of 1,3-propanediol and generally have improved stability and reasonable k _cat values .

Example 14
In this example, the location of mutations with increased T2 or increased T2 / T1 ratio compared to wild-type GDH was examined for three-dimensional crystal construction of GDH. This allows for the identification of regions within the dehydratase that can be considered mutation “hot spots” where mutations often result in improved reaction rates (decreasing inactivation rate). Alternative sequence changes in these regions will likely result in additional mutants with improved reaction rates.

Mutation positions correlating to the three-dimensional structure The three-dimensional crystal structure of the substrate-free form of dehydratase has been determined by X-ray crystal structure analysis (Liao et al., J. Inorganic Biochem. 93 (1-2): 84. 91 (2003)), and the structure of the enzyme in the 1,2-propanediol complex has also been reported (Yamanishi et al., Eur. J. Biochem. 269: 4484-4494 (2002)). Based on these structures, mutations (from Examples 6-12) that resulted in improvements in either the T2 / T1 ratio or T2 were mapped onto the schematic of each GDH subunit. More specifically, FIG. 3 shows the distribution of mutants containing a single point mutation (compared to wild type GDH) on the α-subunit (corresponding to 88% of all single point mutations). In contrast, FIG. 4 shows the distribution of mutants containing multiple point mutations (compared to wild type GDH) on the α-subunit. This subunit contained 58% of all multipoint mutations and the remaining mutations were distributed in β- (31%) and γ- (11%) subunits, respectively.

Hot spots identified from positive hits Mutations in both single-point and multipoint mutants are located in all three subunits of GDH and are distributed throughout the large α-subunit, but contain single-point mutations And mutants containing multiple point mutations show two common hot spots in the α-subunit.

  One mutation hot spot is located on the second α helix (residues 62-70) from the N-terminus of the α-subunit. In this region, 5 single point mutants were found, 3 of which have different amino acid substitution mutations at residue 62, 1 mutation at residue 65 and 1 mutation at residue 70 (FIG. 3). Six mutation sites from the multipoint mutant are also located on this helix, with three at residue 62 and one at residues 63, 67 and 70 (FIG. 4).

  The second hot spot is the region containing the fourth β-strand portion, the TIM barrel followed by the α-subunit loop and a short helix (residues 224-236). This hot spot is in the vicinity of the active site. Five single site mutations have been found in this region. Residue 224 had a different amino acid substitution in the two mutants. Residue 226 is the mutation site of two identical mutants. One mutant is on residue 233. Furthermore, the three mutation sites of the multi-site mutant are located in this region (residues 226, 231 and 236).

Example 15
Site saturation mutagenesis of GDH mutants Site saturation mutagenesis was performed at the three mutation sites present in the mutants characterized in Example 9. Specifically, these sites were γ-Thr53 (found in mutant Xba3009), α-Leu509 (found in mutant Xba3029) and α-Val224 (found in mutant 4BR1001). To prepare the saturation mutagenesis library, 1-E1 and 8-C9 mutants (Examples 9 and 10, respectively) were purified from their host cells and used as templates with the degenerate primers shown in Table 21.

  Using the Stratagene QuickChange site-induced mutagenesis kit from Stratagene, La Jolla, CA (Stratagene, La Jolla, Calif.) According to the manufacturer's instructions, GDH-SM1, GDH-SM2, GDH-SM3, and GDH-SM4 libraries were prepared. Following electroporation of the plasmid into E. coli strain 5K (DE3) (as described in Example 1), 88 mutant colonies from each library were selected using a single point GDH assay. Screening estimated total enzyme turnover in the presence of high concentrations of 1,3-propanediol. The best hit from each screen was then subjected to DNA sequence analysis. The results are shown in the following table. The SEQ ID number of the DNA sequence of the enzyme is provided in the first column of the table.

  All four saturation mutants (GDH-SM1-G11, GSH-SM2-B11, GDH-SM3-D2, and GDH-SM4-H2 [shown in bold]) are The comparison showed a further improvement in T (600). Interestingly, in the mutant GDH-SM4-H2, three base changes were identified (ie ACC (γ-Thr53) to TGT (Cys)). This type of mutation is extremely difficult to generate using error-prone PCR.

Example 16
Genetically modified generation extension method to generate GDH mutants using unpaired primers Glycerol using random mutagenesis, rational design mutagenesis, and saturation mutagenesis (Examples 6, 8-11, and 15) Despite a significant improvement in the rate of GDH inactivation in the presence of 1,3-propanediol, further improvements were desired for industrial applications. Therefore, 24 glycerol dehydratase mutants from Examples 6, 9, 10, and 15 were used as parent templates for single genetically engineered extension reactions using unpaired primer methods (US Patent Application No. 60/360279). Specification).

Attachment of a short lateral DNA fragment to the 5 'or 3' end of the parent gene By PCR, a short lateral DNA fragment is attached to the 5 'or 3' end of the parent gene, followed by genetic recombination using an unpaired primer. Used as a binding site for the developmental extension method. A plasmid containing GDH was purified from host cells and used as a template.

  Specifically, in a standard high fidelity PCR reaction using the Expand High Fidelity PCR System from the Roche Applied Science (Indianapolis, IN), Indianapolis, Indiana. The following template genes were amplified using forward primer GDHM-F1 (SEQ ID NO: 343) and reverse transcription primer GDHM-R1 (SEQ ID NO: 344). 1-E1 (SEQ ID NO: 198), wild type GDH (SEQ ID NO: 1), Xba3023 (SEQ ID NO: 182), Xba3010 (SEQ ID NO: 175), 8-C9 (SEQ ID NO: 212), 4BR1001 (SEQ ID NO: 171), Xba3015 ( SEQ ID NO: 147), Xba3008 (SEQ ID NO: 151), Sma3003 (SEQ ID NO: 143), Xba3016 (SEQ ID NO: 155), and Xba3020 (SEQ ID NO: 159). The resulting PCR product was then mixed together in an equimolar ratio and designated “Mix-1”.

  The following genes were amplified in the same manner using forward primer GDHM-F2 (SEQ ID NO: 345) and reverse transcription primer GDHM-R2 (SEQ ID NO: 346). Xba3007 (SEQ ID NO: 40), Xba3029 (SEQ ID NO: 44), Xba3025 (SEQ ID NO: 52), Sma3009 (SEQ ID NO: 179), Sma3010 (SEQ ID NO: 175), Sma3008 (SEQ ID NO: 151), RsrII001 (SEQ ID NO: 136), PpuMI002 (SEQ ID NO: 128), KG005 (SEQ ID NO: 253), GDH-SM1-G11 (SEQ ID NO: 301), GDH-SM2-B11 (SEQ ID NO: 304), GDH-SM3-D2 (SEQ ID NO: 307), and GDH-SM4 -H2 (SEQ ID NO: 310). The resulting PCR product was then mixed together in an equimolar ratio and designated “Mix-2”.

  The amplified product was purified from agarose gel using a Qiagen DNA extraction kit from Qiagen, Valencia, Calif., And then recombination generation extension with unpaired primers Used as a parent template for.

Generation of genetically engineered mutant products using the unpaired primer method Using the two primers PADH316F1 (SEQ ID NO: 29) and T7T (SEQ ID NO: 30), the genetically modified products from the above 24 parent templates ( That is, a GDH mutant gene) was prepared. By the addition of a short 5 ′ DNA fragment generated by SEQ ID NO: 343, PADH316F1 is converted to “Mix-1” (ie, 1-E1, wild type GDH, Xba3023, Xba3010, 8-C9, 4BR1001, Xba3015, Xba3008, Sma3003). , Xba3016, and Xba3020) with the 5 ′ end of the template. However, primer PADH316F1 does not bind to the 5 ′ end of the “Mix-2” template. Similarly, primer T7T is a “mixture-2” template (ie, Xba3007, Xba3029, Xba3025, Sma3009, Sma3010, Sma3008, RsrII001, PpuMI002, KG005, GDH-SM1-G11, GDH-SM3-B11, GDH-SM3-D , And GDH-SM4-H2), but does not bind to the 3 ′ end of the “Mix-1” template.

The following reaction mixture was assembled for the reaction: 10 ng “Mix-1”, 10 ng “Mix-2”, 200 μM each dNTP, 1 × PCR buffer (final concentration, 1.5 mM MgCl ₂ ), 286 nM pADH316F1, 286 nM T7T, Qiagen, Valencia, CA (Valencia, CA)) adjusted to 25 μL with 0.625 U HotStarTaq, dH ₂ O. The thermal cycling conditions were as follows: Denaturation at 95 ° C for 2 minutes, followed by 60 cycles of 95 ° C for 30 seconds, 1 second per cycle with a gradient between 1 second + 63-69 ° C, final extension at 72 ° C for 7 minutes, and at 4 ° C Retention. An Eppendorf Mastercycler gradient 5331 from Eppendorf Scientific, Inc. (Westbury, NY) was used for the thermal cycling reaction.

  Since the two primers used in the reaction do not match the 5 'and 3' ends of any template at the same time, neither parent template can be amplified during the reaction. However, any recombinant DNA product having both 5 'and 3' ends will be amplified in subsequent thermal cycling. Following the unpaired primer reaction, the reaction mixture was loaded onto an agarose gel. A large amount of 2.7 kB PCR product was obtained from the unpaired primer reaction at about 66-67 ° C. A product of this size was expected because the original parent molecule used as a template itself was about 2.7 kB.

Generation and screening of a genetically generated GDH mutant library A recombinant GDH DNA product (approximately 2.7 kB) was purified from gel using a Qiagen DNA extraction kit, digested with Xba I and Hind III, It was then ligated with the XbaI-HindIII-digested pGD20 vector (Example 1). As described in Example 1, mutant libraries were obtained by transformation of the ligation mixture by electroporation into E. coli strain 5K (DE3). Library size exceeded 300,000 colonies per ligation reaction.

  6,000-7,000 mutant colonies were picked from agarose plates and screened using a single point high throughput screening assay as described in Example 13. Following the primary screen, a put-up hit was confirmed by a follow-up confirmation assay. Briefly, each putative hit was re-assayed in 8 wells. The results from each individual clone were statistically analyzed to obtain an average value and standard deviation of T (600). These results were compared with the wild type GDH enzyme.

  In addition, a two-point assay was used to investigate some hits in more detail (Example 5) and to roughly estimate changes in the initial kinetics and inactivation of these hits.

  Table 23 summarizes the results of these various assays for several genetically engineered GDH mutants.

Different recombinant GDH mutants showed different enzyme kinetics (despite having similar T (600) values), but this merely reflects the differences in the parameters of each assay means. For example, mutants SHGDH24 and SHGDH43 had the same T (600) value, but SHGDH24 only had a slightly improved T1 value compared to wild type, whereas SHGDH43 had a significantly reduced T1. Had. SHGDH24 did not reduce the enzyme's k _cat , whereas SHGDH43 had a significantly reduced k _cat . In both cases, the increased GDH enzyme stability of both SHGDH24 and SHGDH43 enabled an improvement in total enzyme turnover.

SHGDH51 showed the greatest improvement in stability among these mutants, but its mutants did not have the highest total enzyme turnover because their k _cat was greatly reduced. SHGDH37 showed the greatest improvement for T (600) in these mutants, but its T2 value was lower than that of SHGDH12 and SHGDH37 was higher than SHGDH12 for 1,3-propanediol at high concentrations. It was suggested to be resistant.

  In any case, however, the results clearly demonstrated that the creation of new mutants using the unpaired primer method had a significant improvement in GDH stability. Specifically, the best mutant obtained through random mutagenesis (ie, mutant Sma3002; SEQ ID NO: 140) had a T (600) of 3.3. From a rational combination of mutants identified through random mutagenesis and screening, or from a semi-random combination of these mutations using site saturation mutagenesis, the mutation with a maximum T (600) of 4.3 A strain (ie mutant GDH-SM1-G11; SEQ ID NO: 301) was obtained. Therefore, it is concluded that random genetic recombination using a genetic recombination generation extension approach using unpaired primers is a more powerful technique than the rational and semi-random approach described above.

Sequence analysis of mutant genes Plasmid DNA was purified from the recombinant GDH mutants listed in Table 23, and the entire GDH gene was sequenced. Analysis of the mutant strain and subsequent comparison with the original wild type GDH gene sequence indicate that these recombinant mutant genes contain the single base substitution mutations (point mutations) shown in Table 24. The SEQ ID number of the DNA sequence of the enzyme is provided in the first column of the table.

  Interestingly, all mutants (SEQ ID NOs: 313, 316, 319, 322, 325, 328, 331, 334, 337) contain α-β fusion mutations (originally found in Sma3002 and Xba3009 mutants). Thus, [Example 6]), it was suggested that this mutation is very important for improving the T (600) value. SHGDH43 contained mutations from four different parent genes (ie, 4BR1001, 1-E1, Xba3008, and GDH-SM1-G11) in addition to the four newly created mutations. This suggested that at least 4 crossings occurred between 4 different parent genes during genetically engineered PCR. SHGDH25 and SHGDH51, in addition to several newly created mutations, are mutations from three parent genes (ie, Sma3003, 4BR1001, and 1-E1 for SHGDH25; RsrII001, 1-E1, and Xba3007 for SHGDH51) Contained.

2 shows the time course of a typical GDH reaction performed either in the presence of glycerol or in the presence of glycerol and 1,3-propanediol. Figure 7 is a graphical result of a typical follow-up assay as described in Example 6. The distribution of mutants containing a single point mutation on the α-subunit (compared to wild type GDH) is shown. Figure 3 shows the distribution of mutants containing multiple point mutations on the α-subunit (compared to wild type GDH). The principle of the genetic recombination generation extension method using an unpaired primer is shown.

Claims (10)

(C) contacting a B _12- dependent dehydratase holoenzyme with a mixture comprising glycerol and at least 25 mM 1,3-propanediol;
(D) B _12-dependent dehydratase holoenzyme and screening, B ₁₂ comprising to estimate the turnover ratio of the dependent dehydratase, the screening methods of the B _12-dependent dehydratase with improved reaction rate. The screening step of (b)
e) growing cells that do not have a source of coenzyme B ₁₂ , dehydratase reactivator, or endogenous B _12- dependent dehydratase on a glycerol-free fermentable carbon substrate;
f) permeabilizing the cells;
g) adding a mixture of coenzyme B ₁₂ , glycerol, and at least 25 mM 1,3-propanediol to the permeabilized cells of step (b);
h) quantifying the 3-hydroxypropionaldehyde produced in step (c),
The method according to claim 1, wherein the quantification is a measurement selected from the group consisting of T1, T2, and T (600). SEQ ID NO: 40, 44, 48, 52, 56, 60, 64, 68, 72, 76, 80, 84, 88, 92, 96, 100, 104, 108, 112, 116, 120, 124, 128, 132 136, 140, 143, 147, 151, 155, 159, 163, 167, 171, 175, 179, 182, 186, 189, 192, 195, 198, 201, 204, 208, 212, 215, 218, 221 225, 229, 233, 237, 241, 245, 249, 253, 257, 261, 265, 269, 273, 277, 281, 285, 289, 293, 297, 301, 304, 307, 310, 313, 316 319, 322, 325, 328, 331, 334, and 337, selected from the group consisting of B ₁₂ A nucleic acid sequence encoding a viable mutant dehydratase. SEQ ID NO: 40, 44, 48, 52, 56, 60, 64, 68, 72, 76, 80, 84, 88, 92, 96, 100, 104, 108, 112, 116, 120, 124, 128, 132 136, 140, 143, 147, 151, 155, 159, 163, 167, 171, 175, 179, 182, 186, 189, 192, 195, 198, 201, 204, 208, 212, 215, 218, 221 225, 229, 233, 237, 241, 245, 249, 253, 257, 261, 265, 269, 273, 277, 281, 285, 289, 293, 297, 301, 304, 307, 310, 313, 316 A nucleic acid sequence selected from the group consisting of 319, 322, 325, 328, 331, 334, and 337 B _12-dependent mutant dehydratase encoded I. SEQ ID NOs: 140, 179, 186, 189, 192, 195, 198, 201, 212, 215, 218, 301, 304, 307, 310, 313, 316, 319, 322, 325, 328, 331, 334, and It is selected from the group consisting of 337 comprising the alpha-beta subunit fusion, nucleic acid sequences encoding the B _12-dependent mutant dehydratase. SEQ ID NO: 313,322, and is selected from the group consisting of 328 comprising the alpha-beta subunit _fusion, nucleic acid sequences encoding the _{B 12-dependent} mutant dehydratase.   The nucleic acid sequence of claim 5, further comprising a linker sequence between the α and β subunits of the α-β subunit fusion, wherein the linker sequence is selected from the group consisting of SEQ ID NO: 18 and SEQ ID NO: 19. d) contacting the B _12- dependent dehydratase holoenzyme comprising hot spot 1 or hot spot 2 with the causative agent,
e) having an improved reaction rate comprising screening the mutant produced in step A) for improved kcat and / or stability and f) repeating steps a) and b) how to create a B ₁₂ dependent dehydratase mutants. d) contacting the dehydratase holoenzyme with 5-10 mM glycerol and / or 10-300 mM 1,3-propanediol;
e) in a high-throughput assay, measured at at least two time points sufficiently separated to estimate k _cat and total enzyme turnover,
f) comprises selecting the B _12-dependent mutants having reaction rates that are improved compared to the standard dehydratase, identifying B _12-dependent dehydratase having a reaction rate which is improved in comparison with standard dehydratase how to. d) contacting the dehydratase holoenzyme with 5-50 mM glycerol and> 300 mM 1,3-propanediol,
e) in a high-throughput assay, measured at one time point well separated from T0 to estimate total enzyme turnover, and f) B _12- dependent with improved kinetics compared to standard dehydratase methods of identifying comprising selecting a mutant strain, a B ₁₂ dependent dehydratase having a reaction rate that is improved compared to standard dehydratase.

Images